creditAnalysis
This is the credit analysis and prediction of German Credit Data available on https://archive.ics.uci.edu/ml/datasets/Statlog+%28German+Credit+Data%29
## Importing packages
library(tidyverse) ## ── Attaching packages ──────────────────────────── tidyverse 1.2.1 ──## ✔ ggplot2 3.0.0 ✔ purrr 0.2.5
## ✔ tibble 1.4.2 ✔ dplyr 0.7.7
## ✔ tidyr 0.8.1 ✔ stringr 1.3.1
## ✔ readr 1.1.1 ✔ forcats 0.3.0## ── Conflicts ─────────────────────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag() masks stats::lag()library(MASS)##
## Attaching package: 'MASS'## The following object is masked from 'package:dplyr':
##
## selectlibrary(car)## Loading required package: carData##
## Attaching package: 'car'## The following object is masked from 'package:dplyr':
##
## recode## The following object is masked from 'package:purrr':
##
## somelibrary(caret)## Loading required package: lattice##
## Attaching package: 'caret'## The following object is masked from 'package:purrr':
##
## liftlibrary(randomForest)## randomForest 4.6-14## Type rfNews() to see new features/changes/bug fixes.##
## Attaching package: 'randomForest'## The following object is masked from 'package:dplyr':
##
## combine## The following object is masked from 'package:ggplot2':
##
## marginlibrary(ROCR)## Loading required package: gplots##
## Attaching package: 'gplots'## The following object is masked from 'package:stats':
##
## lowesslibrary(e1071)
## Loading data into environment
creditData <- read.table("german.data")
## create proper column names
colnames(creditData) <- c('existingcheckingstatus', 'duration', 'credithistory', 'purpose', 'creditamount',
'savings', 'employmentlength', 'installmentrate', 'marriagesex', 'otherdebtors',
'presentresidencelength', 'property', 'age', 'otherinstallmentplans', 'housing',
'existingcredits', 'job', 'peopleliableno', 'telephone', 'foreignworker', 'classification')
## convert existingcredits to 0,1 except 1,2
creditData <- creditData %>% mutate(existingcredits = ifelse(existingcredits == 1, 0, 1))
## convert classification top 0,1 except 1,2
creditData <- creditData %>% mutate(classification = ifelse(classification == 1, 0, 1))
## existingcheckingstatus
ggplot(creditData, aes(x = creditData$existingcheckingstatus)) + geom_bar(fill = 'coral') + labs(title = 'Status of existing checking account', x = 'existingcheckingstatus', y = 'Count')
## duration
ggplot(creditData, aes(x = creditData$duration)) + geom_histogram(fill = 'lightblue', binwidth = 3) + labs(title = 'Duration in months', x = 'duration', y = 'Count')
boxplot(creditData$duration)
## impute the outliers
quantile(creditData$duration,seq(0,1,0.01))## 0% 1% 2% 3% 4% 5% 6% 7% 8% 9% 10% 11%
## 4.00 6.00 6.00 6.00 6.00 6.00 6.00 6.00 6.00 8.00 9.00 9.00
## 12% 13% 14% 15% 16% 17% 18% 19% 20% 21% 22% 23%
## 9.00 9.00 9.00 10.00 10.00 10.00 11.82 12.00 12.00 12.00 12.00 12.00
## 24% 25% 26% 27% 28% 29% 30% 31% 32% 33% 34% 35%
## 12.00 12.00 12.00 12.00 12.00 12.00 12.00 12.00 12.00 12.00 12.00 12.00
## 36% 37% 38% 39% 40% 41% 42% 43% 44% 45% 46% 47%
## 13.00 15.00 15.00 15.00 15.00 15.00 15.00 15.00 18.00 18.00 18.00 18.00
## 48% 49% 50% 51% 52% 53% 54% 55% 56% 57% 58% 59%
## 18.00 18.00 18.00 18.00 18.00 18.00 18.00 20.00 21.00 21.00 21.00 24.00
## 60% 61% 62% 63% 64% 65% 66% 67% 68% 69% 70% 71%
## 24.00 24.00 24.00 24.00 24.00 24.00 24.00 24.00 24.00 24.00 24.00 24.00
## 72% 73% 74% 75% 76% 77% 78% 79% 80% 81% 82% 83%
## 24.00 24.00 24.00 24.00 24.00 24.46 27.00 30.00 30.00 30.00 30.00 33.51
## 84% 85% 86% 87% 88% 89% 90% 91% 92% 93% 94% 95%
## 36.00 36.00 36.00 36.00 36.00 36.00 36.00 36.00 42.00 42.21 48.00 48.00
## 96% 97% 98% 99% 100%
## 48.00 48.00 48.00 60.00 72.00creditData$duration[which(creditData$duration > 48)] <- 48
## credithistory
ggplot(creditData, aes(x = creditData$credithistory)) + geom_bar(fill = 'coral') + labs(title = 'Credit history', x = 'credithistory', y = 'Count')
## purpose
ggplot(creditData, aes(x = creditData$purpose)) + geom_bar(fill = 'coral') + labs(title = 'Purpose of loan', x = 'purpose', y = 'Count')
## creditamount
ggplot(creditData, aes(x = creditData$creditamount)) + geom_histogram(fill = 'lightblue', binwidth = 500) + labs(title = 'Credit amount', x = 'creditamount', y = 'Count')
## boxplot
boxplot(creditData$creditamount)
## outlier treatment
quantile(creditData$creditamount,seq(0,1,0.01))## 0% 1% 2% 3% 4% 5% 6% 7%
## 250.00 425.83 570.02 638.70 682.96 708.95 744.76 782.86
## 8% 9% 10% 11% 12% 13% 14% 15%
## 845.60 901.82 932.00 973.35 1037.88 1074.96 1121.60 1157.55
## 16% 17% 18% 19% 20% 21% 22% 23%
## 1192.52 1208.66 1230.46 1243.24 1262.00 1282.00 1296.56 1321.08
## 24% 25% 26% 27% 28% 29% 30% 31%
## 1344.76 1365.50 1384.96 1409.00 1424.00 1454.42 1479.40 1509.83
## 32% 33% 34% 35% 36% 37% 38% 39%
## 1533.00 1551.01 1573.32 1602.65 1730.60 1788.71 1832.72 1870.05
## 40% 41% 42% 43% 44% 45% 46% 47%
## 1906.80 1931.54 1951.74 1985.71 2039.00 2100.55 2135.08 2176.30
## 48% 49% 50% 51% 52% 53% 54% 55%
## 2236.56 2278.51 2319.50 2341.41 2389.48 2440.41 2512.84 2578.00
## 56% 57% 58% 59% 60% 61% 62% 63%
## 2626.76 2681.15 2746.84 2780.23 2852.40 2923.39 2997.90 3059.37
## 64% 65% 66% 67% 68% 69% 70% 71%
## 3106.08 3187.40 3334.74 3384.66 3446.32 3529.48 3590.00 3621.29
## 72% 73% 74% 75% 76% 77% 78% 79%
## 3721.64 3832.81 3913.26 3972.25 4110.72 4254.29 4442.30 4591.63
## 80% 81% 82% 83% 84% 85% 86% 87%
## 4720.00 4848.94 5150.36 5389.84 5800.16 5969.95 6224.70 6366.46
## 88% 89% 90% 91% 92% 93% 94% 95%
## 6583.20 6852.42 7179.40 7419.26 7687.88 7985.95 8471.96 9162.70
## 96% 97% 98% 99% 100%
## 9966.68 10961.39 12169.70 14180.39 18424.00creditData$creditamount[which(creditData$creditamount > 9162.70000000001)] <- 9162.70000000001
## savings
ggplot(creditData, aes(x = creditData$savings)) + geom_bar(fill = 'coral') + labs(title = 'Savings account/bonds', x = 'savings', y = 'Count')
## employmentlength
ggplot(creditData, aes(x = creditData$employmentlength)) + geom_bar(fill = 'coral') + labs(title = 'Present employment since', x = 'employmentlength', y = 'Count')
## installmentrate
ggplot(creditData, aes(x = creditData$installmentrate)) + geom_histogram(fill = 'lightblue', binwidth = 1) + labs(title = 'Installment rate in percentage of disposable income', x = 'installmentrate', y = 'Count')
## boxplot
boxplot(creditData$installmentrate)
## no outliers
## marriagesex
ggplot(creditData, aes(x = creditData$marriagesex)) + geom_bar(fill = 'coral') + labs(title = 'Personal status and sex', x = 'marriagesex', y = 'Count')
## otherdebtors
ggplot(creditData, aes(x = creditData$otherdebtors)) + geom_bar(fill = 'coral') + labs(title = 'Other debtors / guarantors', x = 'otherdebtors', y = 'Count')
## almost all the people have not any othe debtors/ guarantorsw
## presentresidencelength
ggplot(creditData, aes(x = creditData$presentresidencelength)) + geom_histogram(fill = 'lightblue', binwidth = 1) + labs(title = 'Present residence since', x = 'presentresidencelength', y = 'Count')
## boxplot
boxplot(creditData$presentresidencelength)
## no outlier
## property
ggplot(creditData, aes(x = creditData$property)) + geom_bar(fill = 'coral') + labs(title = 'Property', x = 'property', y = 'Count')
## age
ggplot(creditData, aes(x = creditData$age)) + geom_histogram(fill = 'lightblue', binwidth = 10) + labs(title = 'Age in years', x = 'age', y = 'Count')
## boxplot
boxplot(creditData$age)
## outlier treatment
quantile(creditData$age,seq(0,1,0.01))## 0% 1% 2% 3% 4% 5% 6% 7% 8% 9% 10% 11%
## 19.00 20.00 21.00 21.97 22.00 22.00 23.00 23.00 23.00 23.00 23.00 24.00
## 12% 13% 14% 15% 16% 17% 18% 19% 20% 21% 22% 23%
## 24.00 24.00 24.00 25.00 25.00 25.00 25.00 25.81 26.00 26.00 26.00 26.00
## 24% 25% 26% 27% 28% 29% 30% 31% 32% 33% 34% 35%
## 26.76 27.00 27.00 27.00 27.00 27.00 28.00 28.00 28.00 28.00 29.00 29.00
## 36% 37% 38% 39% 40% 41% 42% 43% 44% 45% 46% 47%
## 29.00 29.00 30.00 30.00 30.00 30.00 31.00 31.00 31.00 32.00 32.00 32.00
## 48% 49% 50% 51% 52% 53% 54% 55% 56% 57% 58% 59%
## 32.00 33.00 33.00 33.00 34.00 34.00 34.00 35.00 35.00 35.00 35.00 36.00
## 60% 61% 62% 63% 64% 65% 66% 67% 68% 69% 70% 71%
## 36.00 36.00 36.00 37.00 37.00 37.00 38.00 38.00 38.32 39.00 39.00 40.00
## 72% 73% 74% 75% 76% 77% 78% 79% 80% 81% 82% 83%
## 40.00 41.00 41.00 42.00 42.00 43.00 43.00 44.00 45.00 45.00 46.00 46.00
## 84% 85% 86% 87% 88% 89% 90% 91% 92% 93% 94% 95%
## 47.00 48.00 48.00 49.00 50.00 51.00 52.00 53.00 54.00 56.00 57.00 60.00
## 96% 97% 98% 99% 100%
## 61.00 63.00 65.00 67.01 75.00creditData$age[which(creditData$age > 67)] <- 67
## otherinstallmentplans
ggplot(creditData, aes(x = creditData$otherinstallmentplans)) + geom_bar(fill = 'coral') + labs(title = 'Other installment plans', x = 'otherinstallmentplans', y = 'Count')
## none outnumbers all the others
## housing
ggplot(creditData, aes(x = creditData$housing)) + geom_bar(fill = 'coral') + labs(title = 'Housing', x = 'housing', y = 'Count')
## own outnumbers all of the others
## existingcredits
ggplot(creditData, aes(x = creditData$existingcredits)) + geom_histogram(fill = 'lightblue', binwidth = .5) + labs(title = 'Number of existing credits at this bank', x = 'existingcredits', y = 'Count')
## boxplot
boxplot(creditData$existingcredits)
## no outliers
## existing credits as factor
ggplot(creditData, aes(x = as.factor(creditData$existingcredits))) + geom_bar(fill = 'coral') + labs(title = 'Number of existing credits at this bank', x = 'existingcredits', y = 'Count')
## job
ggplot(creditData, aes(x = creditData$job)) + geom_bar(fill = 'coral') + labs(title = 'Job', x = 'job', y = 'Count')
## unemployed/ unskilled - non-resident is vewry low in count
## peopleliableno
ggplot(creditData, aes(x = creditData$peopleliableno)) + geom_histogram(fill = 'lightblue', binwidth = 1) + labs(title = 'Number of people being liable to provide maintenance for', x = 'no.of peopleliable', y = 'Count')
## boxplot
boxplot(creditData$peopleliableno)
## outlier treatment
quantile(creditData$peopleliableno,seq(0,1,0.01))## 0% 1% 2% 3% 4% 5% 6% 7% 8% 9% 10% 11% 12% 13% 14%
## 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
## 15% 16% 17% 18% 19% 20% 21% 22% 23% 24% 25% 26% 27% 28% 29%
## 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
## 30% 31% 32% 33% 34% 35% 36% 37% 38% 39% 40% 41% 42% 43% 44%
## 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
## 45% 46% 47% 48% 49% 50% 51% 52% 53% 54% 55% 56% 57% 58% 59%
## 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
## 60% 61% 62% 63% 64% 65% 66% 67% 68% 69% 70% 71% 72% 73% 74%
## 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
## 75% 76% 77% 78% 79% 80% 81% 82% 83% 84% 85% 86% 87% 88% 89%
## 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2
## 90% 91% 92% 93% 94% 95% 96% 97% 98% 99% 100%
## 2 2 2 2 2 2 2 2 2 2 2# no need
## telephone
ggplot(creditData, aes(x = creditData$telephone)) + geom_bar(fill = 'coral') + labs(title = 'Telephone present', x = 'telephone', y = 'Count')
## foreignworker
ggplot(creditData, aes(x = creditData$foreignworker)) + geom_bar(fill = 'coral') + labs(title = 'Foreign worker', x = 'foreignworker', y = 'Count')
## yes outnumbers no with a large margin
## check status of good and bad
ggplot(creditData, aes(x = as.factor(creditData$classification))) + geom_bar(fill = 'coral') + labs(title = 'Good vs Bad', x = 'classification', y = 'Count')
## age vs credit amount for various purposes
ggplot(creditData, aes(x = age, y = creditamount)) + geom_point(aes(color=as.factor(classification))) + facet_wrap(~purpose)
## age vs employment length
ggplot(creditData, aes(y = age, x = employmentlength, fill = as.factor(classification))) + stat_summary(fun.y="mean", geom="bar", position = 'dodge') 
## not much relation between age and employment length
## employment length vs savings
ggplot(creditData, aes(x = employmentlength)) + geom_bar(aes(fill = savings), position = 'dodge') + facet_wrap(~classification)
## all follow almost similar trends
## convert telephone and foreignworker into 0 and 1, as they are factors with 2 levels
levels(creditData$telephone) <- c(0,1)
creditData$telephone <- as.numeric(levels(creditData$telephone))[creditData$telephone]
levels(creditData$foreignworker) <- c(0,1)
creditData$foreignworker <- as.numeric(levels(creditData$foreignworker))[creditData$foreignworker]
## convert other factores with more than 2 levels into dummies
creditDataNumeric <- as.data.frame(model.matrix(~ ., data = creditData, contrasts.arg = lapply(creditData[,sapply(creditData, is.factor)], contrasts, contrasts=FALSE)))[,-1]
### model creation
## test train split
set.seed(100)
indices = sample(1:nrow(creditDataNumeric), 0.7*nrow(creditDataNumeric))
train = creditDataNumeric[indices,]
test = creditDataNumeric[-indices,]
## penalty matrix
penaltyMatrix <- matrix(c(0,1,5,0), byrow = T, nrow = 2)
## glm model
model1 <- glm(classification~., data = creditDataNumeric)
summary(model1)##
## Call:
## glm(formula = classification ~ ., data = creditDataNumeric)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.91222 -0.28361 -0.09102 0.31827 1.02409
##
## Coefficients: (11 not defined because of singularities)
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -4.417e-01 1.620e-01 -2.727 0.006510 **
## existingcheckingstatusA11 2.689e-01 3.455e-02 7.784 1.84e-14 ***
## existingcheckingstatusA12 1.949e-01 3.375e-02 5.775 1.04e-08 ***
## existingcheckingstatusA13 8.106e-02 5.588e-02 1.450 0.147267
## existingcheckingstatusA14 NA NA NA NA
## duration 5.611e-03 1.630e-03 3.443 0.000601 ***
## credithistoryA30 2.645e-01 7.167e-02 3.690 0.000237 ***
## credithistoryA31 2.715e-01 6.976e-02 3.892 0.000106 ***
## credithistoryA32 1.227e-01 3.803e-02 3.227 0.001293 **
## credithistoryA33 6.900e-02 5.106e-02 1.351 0.176916
## credithistoryA34 NA NA NA NA
## purposeA40 1.218e-01 5.167e-02 2.357 0.018603 *
## purposeA41 -1.161e-01 6.106e-02 -1.901 0.057571 .
## purposeA410 -1.048e-01 1.285e-01 -0.816 0.414723
## purposeA42 -7.880e-03 5.424e-02 -0.145 0.884525
## purposeA43 -1.371e-02 5.066e-02 -0.271 0.786684
## purposeA44 4.186e-02 1.256e-01 0.333 0.738909
## purposeA45 8.203e-02 9.668e-02 0.848 0.396394
## purposeA46 1.418e-01 7.348e-02 1.930 0.053908 .
## purposeA48 -1.343e-01 1.436e-01 -0.935 0.350184
## purposeA49 NA NA NA NA
## creditamount 1.573e-05 8.930e-06 1.761 0.078534 .
## savingsA61 1.246e-01 3.579e-02 3.482 0.000520 ***
## savingsA62 7.424e-02 5.086e-02 1.460 0.144740
## savingsA63 4.252e-02 5.960e-02 0.713 0.475805
## savingsA64 -2.511e-02 6.626e-02 -0.379 0.704824
## savingsA65 NA NA NA NA
## employmentlengthA71 5.143e-02 6.625e-02 0.776 0.437697
## employmentlengthA72 4.675e-02 4.594e-02 1.018 0.309059
## employmentlengthA73 2.255e-02 3.757e-02 0.600 0.548461
## employmentlengthA74 -6.703e-02 4.231e-02 -1.584 0.113493
## employmentlengthA75 NA NA NA NA
## installmentrate 4.138e-02 1.364e-02 3.033 0.002484 **
## marriagesexA91 8.131e-02 7.328e-02 1.110 0.267447
## marriagesexA92 2.907e-02 4.904e-02 0.593 0.553464
## marriagesexA93 -4.337e-02 4.881e-02 -0.889 0.374475
## marriagesexA94 NA NA NA NA
## otherdebtorsA101 1.703e-01 6.067e-02 2.807 0.005101 **
## otherdebtorsA102 2.457e-01 8.669e-02 2.834 0.004688 **
## otherdebtorsA103 NA NA NA NA
## presentresidencelength 2.258e-04 1.325e-02 0.017 0.986411
## propertyA121 -1.260e-01 6.344e-02 -1.986 0.047326 *
## propertyA122 -8.278e-02 6.252e-02 -1.324 0.185821
## propertyA123 -9.555e-02 6.075e-02 -1.573 0.116096
## propertyA124 NA NA NA NA
## age -1.634e-03 1.366e-03 -1.196 0.231822
## otherinstallmentplansA141 8.155e-02 3.902e-02 2.090 0.036899 *
## otherinstallmentplansA142 7.851e-02 6.235e-02 1.259 0.208249
## otherinstallmentplansA143 NA NA NA NA
## housingA151 1.210e-01 7.221e-02 1.676 0.094052 .
## housingA152 5.001e-02 6.889e-02 0.726 0.468046
## housingA153 NA NA NA NA
## existingcredits 5.091e-02 3.504e-02 1.453 0.146572
## jobA171 -7.311e-02 1.024e-01 -0.714 0.475190
## jobA172 -4.067e-03 5.401e-02 -0.075 0.939997
## jobA173 5.963e-03 4.407e-02 0.135 0.892399
## jobA174 NA NA NA NA
## peopleliableno 3.170e-02 3.800e-02 0.834 0.404421
## telephone -4.083e-02 2.963e-02 -1.378 0.168532
## foreignworker -1.403e-01 7.059e-02 -1.988 0.047136 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 0.159032)
##
## Null deviance: 210.00 on 999 degrees of freedom
## Residual deviance: 151.24 on 951 degrees of freedom
## AIC: 1049
##
## Number of Fisher Scoring iterations: 2## uset stepAIC to reduce colinearity
step <- stepAIC(model1)## Start: AIC=1048.99
## classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
## existingcheckingstatusA13 + existingcheckingstatusA14 + duration +
## credithistoryA30 + credithistoryA31 + credithistoryA32 +
## credithistoryA33 + credithistoryA34 + purposeA40 + purposeA41 +
## purposeA410 + purposeA42 + purposeA43 + purposeA44 + purposeA45 +
## purposeA46 + purposeA48 + purposeA49 + creditamount + savingsA61 +
## savingsA62 + savingsA63 + savingsA64 + savingsA65 + employmentlengthA71 +
## employmentlengthA72 + employmentlengthA73 + employmentlengthA74 +
## employmentlengthA75 + installmentrate + marriagesexA91 +
## marriagesexA92 + marriagesexA93 + marriagesexA94 + otherdebtorsA101 +
## otherdebtorsA102 + otherdebtorsA103 + presentresidencelength +
## propertyA121 + propertyA122 + propertyA123 + propertyA124 +
## age + otherinstallmentplansA141 + otherinstallmentplansA142 +
## otherinstallmentplansA143 + housingA151 + housingA152 + housingA153 +
## existingcredits + jobA171 + jobA172 + jobA173 + jobA174 +
## peopleliableno + telephone + foreignworker
##
##
## Step: AIC=1048.99
## classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
## existingcheckingstatusA13 + existingcheckingstatusA14 + duration +
## credithistoryA30 + credithistoryA31 + credithistoryA32 +
## credithistoryA33 + credithistoryA34 + purposeA40 + purposeA41 +
## purposeA410 + purposeA42 + purposeA43 + purposeA44 + purposeA45 +
## purposeA46 + purposeA48 + purposeA49 + creditamount + savingsA61 +
## savingsA62 + savingsA63 + savingsA64 + savingsA65 + employmentlengthA71 +
## employmentlengthA72 + employmentlengthA73 + employmentlengthA74 +
## employmentlengthA75 + installmentrate + marriagesexA91 +
## marriagesexA92 + marriagesexA93 + marriagesexA94 + otherdebtorsA101 +
## otherdebtorsA102 + otherdebtorsA103 + presentresidencelength +
## propertyA121 + propertyA122 + propertyA123 + propertyA124 +
## age + otherinstallmentplansA141 + otherinstallmentplansA142 +
## otherinstallmentplansA143 + housingA151 + housingA152 + housingA153 +
## existingcredits + jobA171 + jobA172 + jobA173 + peopleliableno +
## telephone + foreignworker
##
##
## Step: AIC=1048.99
## classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
## existingcheckingstatusA13 + existingcheckingstatusA14 + duration +
## credithistoryA30 + credithistoryA31 + credithistoryA32 +
## credithistoryA33 + credithistoryA34 + purposeA40 + purposeA41 +
## purposeA410 + purposeA42 + purposeA43 + purposeA44 + purposeA45 +
## purposeA46 + purposeA48 + purposeA49 + creditamount + savingsA61 +
## savingsA62 + savingsA63 + savingsA64 + savingsA65 + employmentlengthA71 +
## employmentlengthA72 + employmentlengthA73 + employmentlengthA74 +
## employmentlengthA75 + installmentrate + marriagesexA91 +
## marriagesexA92 + marriagesexA93 + marriagesexA94 + otherdebtorsA101 +
## otherdebtorsA102 + otherdebtorsA103 + presentresidencelength +
## propertyA121 + propertyA122 + propertyA123 + propertyA124 +
## age + otherinstallmentplansA141 + otherinstallmentplansA142 +
## otherinstallmentplansA143 + housingA151 + housingA152 + existingcredits +
## jobA171 + jobA172 + jobA173 + peopleliableno + telephone +
## foreignworker
##
##
## Step: AIC=1048.99
## classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
## existingcheckingstatusA13 + existingcheckingstatusA14 + duration +
## credithistoryA30 + credithistoryA31 + credithistoryA32 +
## credithistoryA33 + credithistoryA34 + purposeA40 + purposeA41 +
## purposeA410 + purposeA42 + purposeA43 + purposeA44 + purposeA45 +
## purposeA46 + purposeA48 + purposeA49 + creditamount + savingsA61 +
## savingsA62 + savingsA63 + savingsA64 + savingsA65 + employmentlengthA71 +
## employmentlengthA72 + employmentlengthA73 + employmentlengthA74 +
## employmentlengthA75 + installmentrate + marriagesexA91 +
## marriagesexA92 + marriagesexA93 + marriagesexA94 + otherdebtorsA101 +
## otherdebtorsA102 + otherdebtorsA103 + presentresidencelength +
## propertyA121 + propertyA122 + propertyA123 + propertyA124 +
## age + otherinstallmentplansA141 + otherinstallmentplansA142 +
## housingA151 + housingA152 + existingcredits + jobA171 + jobA172 +
## jobA173 + peopleliableno + telephone + foreignworker
##
##
## Step: AIC=1048.99
## classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
## existingcheckingstatusA13 + existingcheckingstatusA14 + duration +
## credithistoryA30 + credithistoryA31 + credithistoryA32 +
## credithistoryA33 + credithistoryA34 + purposeA40 + purposeA41 +
## purposeA410 + purposeA42 + purposeA43 + purposeA44 + purposeA45 +
## purposeA46 + purposeA48 + purposeA49 + creditamount + savingsA61 +
## savingsA62 + savingsA63 + savingsA64 + savingsA65 + employmentlengthA71 +
## employmentlengthA72 + employmentlengthA73 + employmentlengthA74 +
## employmentlengthA75 + installmentrate + marriagesexA91 +
## marriagesexA92 + marriagesexA93 + marriagesexA94 + otherdebtorsA101 +
## otherdebtorsA102 + otherdebtorsA103 + presentresidencelength +
## propertyA121 + propertyA122 + propertyA123 + age + otherinstallmentplansA141 +
## otherinstallmentplansA142 + housingA151 + housingA152 + existingcredits +
## jobA171 + jobA172 + jobA173 + peopleliableno + telephone +
## foreignworker
##
##
## Step: AIC=1048.99
## classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
## existingcheckingstatusA13 + existingcheckingstatusA14 + duration +
## credithistoryA30 + credithistoryA31 + credithistoryA32 +
## credithistoryA33 + credithistoryA34 + purposeA40 + purposeA41 +
## purposeA410 + purposeA42 + purposeA43 + purposeA44 + purposeA45 +
## purposeA46 + purposeA48 + purposeA49 + creditamount + savingsA61 +
## savingsA62 + savingsA63 + savingsA64 + savingsA65 + employmentlengthA71 +
## employmentlengthA72 + employmentlengthA73 + employmentlengthA74 +
## employmentlengthA75 + installmentrate + marriagesexA91 +
## marriagesexA92 + marriagesexA93 + marriagesexA94 + otherdebtorsA101 +
## otherdebtorsA102 + presentresidencelength + propertyA121 +
## propertyA122 + propertyA123 + age + otherinstallmentplansA141 +
## otherinstallmentplansA142 + housingA151 + housingA152 + existingcredits +
## jobA171 + jobA172 + jobA173 + peopleliableno + telephone +
## foreignworker
##
##
## Step: AIC=1048.99
## classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
## existingcheckingstatusA13 + existingcheckingstatusA14 + duration +
## credithistoryA30 + credithistoryA31 + credithistoryA32 +
## credithistoryA33 + credithistoryA34 + purposeA40 + purposeA41 +
## purposeA410 + purposeA42 + purposeA43 + purposeA44 + purposeA45 +
## purposeA46 + purposeA48 + purposeA49 + creditamount + savingsA61 +
## savingsA62 + savingsA63 + savingsA64 + savingsA65 + employmentlengthA71 +
## employmentlengthA72 + employmentlengthA73 + employmentlengthA74 +
## employmentlengthA75 + installmentrate + marriagesexA91 +
## marriagesexA92 + marriagesexA93 + otherdebtorsA101 + otherdebtorsA102 +
## presentresidencelength + propertyA121 + propertyA122 + propertyA123 +
## age + otherinstallmentplansA141 + otherinstallmentplansA142 +
## housingA151 + housingA152 + existingcredits + jobA171 + jobA172 +
## jobA173 + peopleliableno + telephone + foreignworker
##
##
## Step: AIC=1048.99
## classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
## existingcheckingstatusA13 + existingcheckingstatusA14 + duration +
## credithistoryA30 + credithistoryA31 + credithistoryA32 +
## credithistoryA33 + credithistoryA34 + purposeA40 + purposeA41 +
## purposeA410 + purposeA42 + purposeA43 + purposeA44 + purposeA45 +
## purposeA46 + purposeA48 + purposeA49 + creditamount + savingsA61 +
## savingsA62 + savingsA63 + savingsA64 + savingsA65 + employmentlengthA71 +
## employmentlengthA72 + employmentlengthA73 + employmentlengthA74 +
## installmentrate + marriagesexA91 + marriagesexA92 + marriagesexA93 +
## otherdebtorsA101 + otherdebtorsA102 + presentresidencelength +
## propertyA121 + propertyA122 + propertyA123 + age + otherinstallmentplansA141 +
## otherinstallmentplansA142 + housingA151 + housingA152 + existingcredits +
## jobA171 + jobA172 + jobA173 + peopleliableno + telephone +
## foreignworker
##
##
## Step: AIC=1048.99
## classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
## existingcheckingstatusA13 + existingcheckingstatusA14 + duration +
## credithistoryA30 + credithistoryA31 + credithistoryA32 +
## credithistoryA33 + credithistoryA34 + purposeA40 + purposeA41 +
## purposeA410 + purposeA42 + purposeA43 + purposeA44 + purposeA45 +
## purposeA46 + purposeA48 + purposeA49 + creditamount + savingsA61 +
## savingsA62 + savingsA63 + savingsA64 + employmentlengthA71 +
## employmentlengthA72 + employmentlengthA73 + employmentlengthA74 +
## installmentrate + marriagesexA91 + marriagesexA92 + marriagesexA93 +
## otherdebtorsA101 + otherdebtorsA102 + presentresidencelength +
## propertyA121 + propertyA122 + propertyA123 + age + otherinstallmentplansA141 +
## otherinstallmentplansA142 + housingA151 + housingA152 + existingcredits +
## jobA171 + jobA172 + jobA173 + peopleliableno + telephone +
## foreignworker
##
##
## Step: AIC=1048.99
## classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
## existingcheckingstatusA13 + existingcheckingstatusA14 + duration +
## credithistoryA30 + credithistoryA31 + credithistoryA32 +
## credithistoryA33 + credithistoryA34 + purposeA40 + purposeA41 +
## purposeA410 + purposeA42 + purposeA43 + purposeA44 + purposeA45 +
## purposeA46 + purposeA48 + creditamount + savingsA61 + savingsA62 +
## savingsA63 + savingsA64 + employmentlengthA71 + employmentlengthA72 +
## employmentlengthA73 + employmentlengthA74 + installmentrate +
## marriagesexA91 + marriagesexA92 + marriagesexA93 + otherdebtorsA101 +
## otherdebtorsA102 + presentresidencelength + propertyA121 +
## propertyA122 + propertyA123 + age + otherinstallmentplansA141 +
## otherinstallmentplansA142 + housingA151 + housingA152 + existingcredits +
## jobA171 + jobA172 + jobA173 + peopleliableno + telephone +
## foreignworker
##
##
## Step: AIC=1048.99
## classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
## existingcheckingstatusA13 + existingcheckingstatusA14 + duration +
## credithistoryA30 + credithistoryA31 + credithistoryA32 +
## credithistoryA33 + purposeA40 + purposeA41 + purposeA410 +
## purposeA42 + purposeA43 + purposeA44 + purposeA45 + purposeA46 +
## purposeA48 + creditamount + savingsA61 + savingsA62 + savingsA63 +
## savingsA64 + employmentlengthA71 + employmentlengthA72 +
## employmentlengthA73 + employmentlengthA74 + installmentrate +
## marriagesexA91 + marriagesexA92 + marriagesexA93 + otherdebtorsA101 +
## otherdebtorsA102 + presentresidencelength + propertyA121 +
## propertyA122 + propertyA123 + age + otherinstallmentplansA141 +
## otherinstallmentplansA142 + housingA151 + housingA152 + existingcredits +
## jobA171 + jobA172 + jobA173 + peopleliableno + telephone +
## foreignworker
##
##
## Step: AIC=1048.99
## classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
## existingcheckingstatusA13 + duration + credithistoryA30 +
## credithistoryA31 + credithistoryA32 + credithistoryA33 +
## purposeA40 + purposeA41 + purposeA410 + purposeA42 + purposeA43 +
## purposeA44 + purposeA45 + purposeA46 + purposeA48 + creditamount +
## savingsA61 + savingsA62 + savingsA63 + savingsA64 + employmentlengthA71 +
## employmentlengthA72 + employmentlengthA73 + employmentlengthA74 +
## installmentrate + marriagesexA91 + marriagesexA92 + marriagesexA93 +
## otherdebtorsA101 + otherdebtorsA102 + presentresidencelength +
## propertyA121 + propertyA122 + propertyA123 + age + otherinstallmentplansA141 +
## otherinstallmentplansA142 + housingA151 + housingA152 + existingcredits +
## jobA171 + jobA172 + jobA173 + peopleliableno + telephone +
## foreignworker
##
## Df Deviance AIC
## - presentresidencelength 1 151.24 1047.0
## - jobA172 1 151.24 1047.0
## - jobA173 1 151.24 1047.0
## - purposeA42 1 151.24 1047.0
## - purposeA43 1 151.25 1047.1
## - purposeA44 1 151.26 1047.1
## - savingsA64 1 151.26 1047.1
## - marriagesexA92 1 151.29 1047.4
## - employmentlengthA73 1 151.30 1047.4
## - savingsA63 1 151.32 1047.5
## - jobA171 1 151.32 1047.5
## - housingA152 1 151.32 1047.5
## - employmentlengthA71 1 151.34 1047.6
## - purposeA410 1 151.34 1047.7
## - peopleliableno 1 151.35 1047.7
## - purposeA45 1 151.35 1047.7
## - marriagesexA93 1 151.37 1047.8
## - purposeA48 1 151.38 1047.9
## - employmentlengthA72 1 151.40 1048.1
## - marriagesexA91 1 151.44 1048.3
## - age 1 151.47 1048.5
## - otherinstallmentplansA142 1 151.49 1048.7
## - propertyA122 1 151.52 1048.8
## - credithistoryA33 1 151.53 1048.9
## - telephone 1 151.54 1049.0
## <none> 151.24 1049.0
## - existingcheckingstatusA13 1 151.57 1049.2
## - existingcredits 1 151.57 1049.2
## - savingsA62 1 151.58 1049.2
## - propertyA123 1 151.63 1049.6
## - employmentlengthA74 1 151.64 1049.6
## - housingA151 1 151.69 1049.9
## - creditamount 1 151.73 1050.2
## - purposeA41 1 151.81 1050.8
## - purposeA46 1 151.83 1050.9
## - propertyA121 1 151.87 1051.1
## - foreignworker 1 151.87 1051.1
## - otherinstallmentplansA141 1 151.93 1051.6
## - purposeA40 1 152.12 1052.8
## - otherdebtorsA101 1 152.49 1055.2
## - otherdebtorsA102 1 152.52 1055.4
## - installmentrate 1 152.70 1056.6
## - credithistoryA32 1 152.90 1057.9
## - duration 1 153.12 1059.4
## - savingsA61 1 153.17 1059.7
## - credithistoryA30 1 153.41 1061.2
## - credithistoryA31 1 153.65 1062.8
## - existingcheckingstatusA12 1 156.54 1081.5
## - existingcheckingstatusA11 1 160.87 1108.7
##
## Step: AIC=1046.99
## classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
## existingcheckingstatusA13 + duration + credithistoryA30 +
## credithistoryA31 + credithistoryA32 + credithistoryA33 +
## purposeA40 + purposeA41 + purposeA410 + purposeA42 + purposeA43 +
## purposeA44 + purposeA45 + purposeA46 + purposeA48 + creditamount +
## savingsA61 + savingsA62 + savingsA63 + savingsA64 + employmentlengthA71 +
## employmentlengthA72 + employmentlengthA73 + employmentlengthA74 +
## installmentrate + marriagesexA91 + marriagesexA92 + marriagesexA93 +
## otherdebtorsA101 + otherdebtorsA102 + propertyA121 + propertyA122 +
## propertyA123 + age + otherinstallmentplansA141 + otherinstallmentplansA142 +
## housingA151 + housingA152 + existingcredits + jobA171 + jobA172 +
## jobA173 + peopleliableno + telephone + foreignworker
##
## Df Deviance AIC
## - jobA172 1 151.24 1045.0
## - jobA173 1 151.24 1045.0
## - purposeA42 1 151.24 1045.0
## - purposeA43 1 151.25 1045.1
## - purposeA44 1 151.26 1045.1
## - savingsA64 1 151.26 1045.1
## - marriagesexA92 1 151.30 1045.4
## - employmentlengthA73 1 151.30 1045.4
## - savingsA63 1 151.32 1045.5
## - jobA171 1 151.32 1045.5
## - housingA152 1 151.32 1045.5
## - employmentlengthA71 1 151.34 1045.6
## - purposeA410 1 151.34 1045.7
## - peopleliableno 1 151.35 1045.7
## - purposeA45 1 151.35 1045.8
## - marriagesexA93 1 151.37 1045.8
## - purposeA48 1 151.38 1045.9
## - employmentlengthA72 1 151.41 1046.1
## - marriagesexA91 1 151.44 1046.3
## - age 1 151.47 1046.5
## - otherinstallmentplansA142 1 151.49 1046.7
## - propertyA122 1 151.52 1046.8
## - credithistoryA33 1 151.53 1046.9
## <none> 151.24 1047.0
## - telephone 1 151.54 1047.0
## - existingcheckingstatusA13 1 151.57 1047.2
## - existingcredits 1 151.57 1047.2
## - savingsA62 1 151.58 1047.2
## - propertyA123 1 151.63 1047.6
## - employmentlengthA74 1 151.65 1047.7
## - housingA151 1 151.69 1047.9
## - creditamount 1 151.73 1048.2
## - purposeA41 1 151.82 1048.8
## - purposeA46 1 151.83 1048.9
## - propertyA121 1 151.87 1049.1
## - foreignworker 1 151.87 1049.1
## - otherinstallmentplansA141 1 151.93 1049.6
## - purposeA40 1 152.12 1050.8
## - otherdebtorsA101 1 152.49 1053.2
## - otherdebtorsA102 1 152.52 1053.4
## - installmentrate 1 152.70 1054.6
## - credithistoryA32 1 152.90 1055.9
## - duration 1 153.13 1057.4
## - savingsA61 1 153.17 1057.7
## - credithistoryA30 1 153.41 1059.2
## - credithistoryA31 1 153.65 1060.8
## - existingcheckingstatusA12 1 156.54 1079.5
## - existingcheckingstatusA11 1 160.91 1107.0
##
## Step: AIC=1044.99
## classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
## existingcheckingstatusA13 + duration + credithistoryA30 +
## credithistoryA31 + credithistoryA32 + credithistoryA33 +
## purposeA40 + purposeA41 + purposeA410 + purposeA42 + purposeA43 +
## purposeA44 + purposeA45 + purposeA46 + purposeA48 + creditamount +
## savingsA61 + savingsA62 + savingsA63 + savingsA64 + employmentlengthA71 +
## employmentlengthA72 + employmentlengthA73 + employmentlengthA74 +
## installmentrate + marriagesexA91 + marriagesexA92 + marriagesexA93 +
## otherdebtorsA101 + otherdebtorsA102 + propertyA121 + propertyA122 +
## propertyA123 + age + otherinstallmentplansA141 + otherinstallmentplansA142 +
## housingA151 + housingA152 + existingcredits + jobA171 + jobA173 +
## peopleliableno + telephone + foreignworker
##
## Df Deviance AIC
## - purposeA42 1 151.24 1043.0
## - purposeA43 1 151.25 1043.1
## - jobA173 1 151.25 1043.1
## - purposeA44 1 151.26 1043.1
## - savingsA64 1 151.26 1043.1
## - marriagesexA92 1 151.30 1043.4
## - employmentlengthA73 1 151.30 1043.4
## - savingsA63 1 151.32 1043.5
## - jobA171 1 151.32 1043.5
## - housingA152 1 151.32 1043.5
## - employmentlengthA71 1 151.34 1043.7
## - purposeA410 1 151.34 1043.7
## - peopleliableno 1 151.35 1043.7
## - purposeA45 1 151.35 1043.8
## - marriagesexA93 1 151.37 1043.8
## - purposeA48 1 151.38 1043.9
## - employmentlengthA72 1 151.41 1044.1
## - marriagesexA91 1 151.44 1044.3
## - age 1 151.48 1044.5
## - otherinstallmentplansA142 1 151.49 1044.7
## - propertyA122 1 151.52 1044.8
## - credithistoryA33 1 151.53 1044.9
## <none> 151.24 1045.0
## - telephone 1 151.57 1045.2
## - existingcredits 1 151.57 1045.2
## - existingcheckingstatusA13 1 151.58 1045.2
## - savingsA62 1 151.58 1045.2
## - propertyA123 1 151.63 1045.6
## - employmentlengthA74 1 151.65 1045.7
## - housingA151 1 151.69 1045.9
## - creditamount 1 151.76 1046.4
## - purposeA41 1 151.82 1046.8
## - purposeA46 1 151.83 1046.9
## - foreignworker 1 151.87 1047.1
## - propertyA121 1 151.88 1047.2
## - otherinstallmentplansA141 1 151.94 1047.6
## - purposeA40 1 152.13 1048.8
## - otherdebtorsA101 1 152.50 1051.3
## - otherdebtorsA102 1 152.52 1051.4
## - installmentrate 1 152.75 1052.9
## - credithistoryA32 1 152.90 1053.9
## - duration 1 153.13 1055.4
## - savingsA61 1 153.19 1055.8
## - credithistoryA30 1 153.41 1057.2
## - credithistoryA31 1 153.65 1058.8
## - existingcheckingstatusA12 1 156.54 1077.5
## - existingcheckingstatusA11 1 160.92 1105.0
##
## Step: AIC=1043.01
## classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
## existingcheckingstatusA13 + duration + credithistoryA30 +
## credithistoryA31 + credithistoryA32 + credithistoryA33 +
## purposeA40 + purposeA41 + purposeA410 + purposeA43 + purposeA44 +
## purposeA45 + purposeA46 + purposeA48 + creditamount + savingsA61 +
## savingsA62 + savingsA63 + savingsA64 + employmentlengthA71 +
## employmentlengthA72 + employmentlengthA73 + employmentlengthA74 +
## installmentrate + marriagesexA91 + marriagesexA92 + marriagesexA93 +
## otherdebtorsA101 + otherdebtorsA102 + propertyA121 + propertyA122 +
## propertyA123 + age + otherinstallmentplansA141 + otherinstallmentplansA142 +
## housingA151 + housingA152 + existingcredits + jobA171 + jobA173 +
## peopleliableno + telephone + foreignworker
##
## Df Deviance AIC
## - purposeA43 1 151.25 1041.1
## - jobA173 1 151.26 1041.1
## - savingsA64 1 151.27 1041.2
## - purposeA44 1 151.27 1041.2
## - marriagesexA92 1 151.30 1041.4
## - employmentlengthA73 1 151.30 1041.4
## - savingsA63 1 151.32 1041.5
## - jobA171 1 151.32 1041.5
## - housingA152 1 151.33 1041.6
## - purposeA410 1 151.35 1041.7
## - employmentlengthA71 1 151.35 1041.7
## - peopleliableno 1 151.35 1041.8
## - marriagesexA93 1 151.37 1041.9
## - purposeA48 1 151.38 1041.9
## - purposeA45 1 151.39 1042.0
## - employmentlengthA72 1 151.41 1042.1
## - marriagesexA91 1 151.44 1042.3
## - age 1 151.48 1042.5
## - otherinstallmentplansA142 1 151.50 1042.7
## - propertyA122 1 151.53 1042.9
## <none> 151.24 1043.0
## - credithistoryA33 1 151.55 1043.0
## - telephone 1 151.57 1043.2
## - existingcheckingstatusA13 1 151.58 1043.2
## - existingcredits 1 151.58 1043.3
## - savingsA62 1 151.59 1043.3
## - propertyA123 1 151.64 1043.6
## - employmentlengthA74 1 151.66 1043.7
## - housingA151 1 151.69 1044.0
## - creditamount 1 151.76 1044.4
## - foreignworker 1 151.87 1045.2
## - propertyA121 1 151.88 1045.2
## - otherinstallmentplansA141 1 151.94 1045.6
## - purposeA41 1 152.04 1046.2
## - purposeA46 1 152.09 1046.6
## - otherdebtorsA101 1 152.50 1049.3
## - otherdebtorsA102 1 152.52 1049.4
## - installmentrate 1 152.75 1050.9
## - credithistoryA32 1 152.91 1052.0
## - purposeA40 1 153.10 1053.2
## - duration 1 153.18 1053.7
## - savingsA61 1 153.19 1053.8
## - credithistoryA30 1 153.49 1055.7
## - credithistoryA31 1 153.67 1056.9
## - existingcheckingstatusA12 1 156.56 1075.6
## - existingcheckingstatusA11 1 160.96 1103.3
##
## Step: AIC=1041.07
## classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
## existingcheckingstatusA13 + duration + credithistoryA30 +
## credithistoryA31 + credithistoryA32 + credithistoryA33 +
## purposeA40 + purposeA41 + purposeA410 + purposeA44 + purposeA45 +
## purposeA46 + purposeA48 + creditamount + savingsA61 + savingsA62 +
## savingsA63 + savingsA64 + employmentlengthA71 + employmentlengthA72 +
## employmentlengthA73 + employmentlengthA74 + installmentrate +
## marriagesexA91 + marriagesexA92 + marriagesexA93 + otherdebtorsA101 +
## otherdebtorsA102 + propertyA121 + propertyA122 + propertyA123 +
## age + otherinstallmentplansA141 + otherinstallmentplansA142 +
## housingA151 + housingA152 + existingcredits + jobA171 + jobA173 +
## peopleliableno + telephone + foreignworker
##
## Df Deviance AIC
## - jobA173 1 151.27 1039.2
## - savingsA64 1 151.27 1039.2
## - purposeA44 1 151.28 1039.3
## - marriagesexA92 1 151.31 1039.5
## - employmentlengthA73 1 151.31 1039.5
## - savingsA63 1 151.33 1039.6
## - jobA171 1 151.33 1039.6
## - housingA152 1 151.34 1039.6
## - purposeA410 1 151.35 1039.7
## - employmentlengthA71 1 151.36 1039.8
## - peopleliableno 1 151.36 1039.8
## - marriagesexA93 1 151.38 1039.9
## - purposeA48 1 151.38 1039.9
## - purposeA45 1 151.42 1040.2
## - employmentlengthA72 1 151.42 1040.2
## - marriagesexA91 1 151.47 1040.5
## - age 1 151.49 1040.6
## - otherinstallmentplansA142 1 151.51 1040.8
## - propertyA122 1 151.53 1040.9
## <none> 151.25 1041.1
## - credithistoryA33 1 151.56 1041.1
## - telephone 1 151.57 1041.2
## - existingcheckingstatusA13 1 151.59 1041.3
## - existingcredits 1 151.59 1041.3
## - savingsA62 1 151.60 1041.4
## - propertyA123 1 151.65 1041.7
## - employmentlengthA74 1 151.66 1041.8
## - housingA151 1 151.71 1042.1
## - creditamount 1 151.77 1042.5
## - foreignworker 1 151.88 1043.2
## - propertyA121 1 151.89 1043.3
## - otherinstallmentplansA141 1 151.96 1043.7
## - purposeA41 1 152.08 1044.5
## - purposeA46 1 152.22 1045.5
## - otherdebtorsA101 1 152.55 1047.6
## - otherdebtorsA102 1 152.56 1047.6
## - installmentrate 1 152.75 1048.9
## - credithistoryA32 1 152.91 1050.0
## - duration 1 153.19 1051.8
## - savingsA61 1 153.22 1052.0
## - credithistoryA30 1 153.54 1054.0
## - credithistoryA31 1 153.69 1055.0
## - purposeA40 1 153.83 1056.0
## - existingcheckingstatusA12 1 156.59 1073.7
## - existingcheckingstatusA11 1 161.07 1102.0
##
## Step: AIC=1039.16
## classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
## existingcheckingstatusA13 + duration + credithistoryA30 +
## credithistoryA31 + credithistoryA32 + credithistoryA33 +
## purposeA40 + purposeA41 + purposeA410 + purposeA44 + purposeA45 +
## purposeA46 + purposeA48 + creditamount + savingsA61 + savingsA62 +
## savingsA63 + savingsA64 + employmentlengthA71 + employmentlengthA72 +
## employmentlengthA73 + employmentlengthA74 + installmentrate +
## marriagesexA91 + marriagesexA92 + marriagesexA93 + otherdebtorsA101 +
## otherdebtorsA102 + propertyA121 + propertyA122 + propertyA123 +
## age + otherinstallmentplansA141 + otherinstallmentplansA142 +
## housingA151 + housingA152 + existingcredits + jobA171 + peopleliableno +
## telephone + foreignworker
##
## Df Deviance AIC
## - savingsA64 1 151.29 1037.3
## - purposeA44 1 151.30 1037.4
## - marriagesexA92 1 151.32 1037.5
## - employmentlengthA73 1 151.33 1037.6
## - savingsA63 1 151.34 1037.7
## - housingA152 1 151.35 1037.7
## - jobA171 1 151.36 1037.8
## - employmentlengthA71 1 151.36 1037.8
## - peopleliableno 1 151.37 1037.8
## - purposeA410 1 151.37 1037.8
## - marriagesexA93 1 151.39 1038.0
## - purposeA48 1 151.40 1038.1
## - purposeA45 1 151.43 1038.3
## - employmentlengthA72 1 151.44 1038.3
## - marriagesexA91 1 151.48 1038.6
## - age 1 151.52 1038.8
## - otherinstallmentplansA142 1 151.52 1038.8
## - propertyA122 1 151.54 1039.0
## <none> 151.27 1039.2
## - credithistoryA33 1 151.57 1039.2
## - telephone 1 151.59 1039.3
## - existingcheckingstatusA13 1 151.59 1039.3
## - savingsA62 1 151.61 1039.4
## - existingcredits 1 151.61 1039.4
## - propertyA123 1 151.66 1039.7
## - employmentlengthA74 1 151.68 1039.9
## - housingA151 1 151.71 1040.1
## - creditamount 1 151.78 1040.5
## - foreignworker 1 151.90 1041.3
## - propertyA121 1 151.91 1041.4
## - otherinstallmentplansA141 1 151.96 1041.8
## - purposeA41 1 152.11 1042.7
## - purposeA46 1 152.24 1043.5
## - otherdebtorsA101 1 152.55 1045.6
## - otherdebtorsA102 1 152.57 1045.7
## - installmentrate 1 152.75 1046.9
## - credithistoryA32 1 152.93 1048.1
## - savingsA61 1 153.22 1050.0
## - duration 1 153.26 1050.3
## - credithistoryA30 1 153.56 1052.2
## - credithistoryA31 1 153.69 1053.1
## - purposeA40 1 153.84 1054.0
## - existingcheckingstatusA12 1 156.59 1071.7
## - existingcheckingstatusA11 1 161.07 1100.0
##
## Step: AIC=1037.29
## classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
## existingcheckingstatusA13 + duration + credithistoryA30 +
## credithistoryA31 + credithistoryA32 + credithistoryA33 +
## purposeA40 + purposeA41 + purposeA410 + purposeA44 + purposeA45 +
## purposeA46 + purposeA48 + creditamount + savingsA61 + savingsA62 +
## savingsA63 + employmentlengthA71 + employmentlengthA72 +
## employmentlengthA73 + employmentlengthA74 + installmentrate +
## marriagesexA91 + marriagesexA92 + marriagesexA93 + otherdebtorsA101 +
## otherdebtorsA102 + propertyA121 + propertyA122 + propertyA123 +
## age + otherinstallmentplansA141 + otherinstallmentplansA142 +
## housingA151 + housingA152 + existingcredits + jobA171 + peopleliableno +
## telephone + foreignworker
##
## Df Deviance AIC
## - purposeA44 1 151.32 1035.5
## - marriagesexA92 1 151.34 1035.7
## - employmentlengthA73 1 151.34 1035.7
## - housingA152 1 151.37 1035.8
## - jobA171 1 151.38 1035.9
## - employmentlengthA71 1 151.38 1036.0
## - savingsA63 1 151.39 1036.0
## - purposeA410 1 151.39 1036.0
## - peopleliableno 1 151.39 1036.0
## - marriagesexA93 1 151.41 1036.1
## - purposeA48 1 151.42 1036.2
## - purposeA45 1 151.45 1036.4
## - employmentlengthA72 1 151.45 1036.4
## - marriagesexA91 1 151.49 1036.7
## - otherinstallmentplansA142 1 151.53 1036.9
## - age 1 151.54 1037.0
## - propertyA122 1 151.56 1037.1
## <none> 151.29 1037.3
## - credithistoryA33 1 151.60 1037.3
## - telephone 1 151.61 1037.4
## - existingcheckingstatusA13 1 151.61 1037.5
## - existingcredits 1 151.62 1037.5
## - propertyA123 1 151.67 1037.8
## - employmentlengthA74 1 151.70 1038.0
## - savingsA62 1 151.71 1038.1
## - housingA151 1 151.73 1038.2
## - creditamount 1 151.81 1038.7
## - foreignworker 1 151.92 1039.5
## - propertyA121 1 151.93 1039.5
## - otherinstallmentplansA141 1 151.98 1039.9
## - purposeA41 1 152.13 1040.9
## - purposeA46 1 152.26 1041.7
## - otherdebtorsA101 1 152.56 1043.7
## - otherdebtorsA102 1 152.58 1043.8
## - installmentrate 1 152.77 1045.0
## - credithistoryA32 1 152.96 1046.3
## - duration 1 153.28 1048.4
## - credithistoryA30 1 153.60 1050.5
## - credithistoryA31 1 153.71 1051.2
## - savingsA61 1 153.78 1051.6
## - purposeA40 1 153.86 1052.2
## - existingcheckingstatusA12 1 156.59 1069.8
## - existingcheckingstatusA11 1 161.09 1098.1
##
## Step: AIC=1035.52
## classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
## existingcheckingstatusA13 + duration + credithistoryA30 +
## credithistoryA31 + credithistoryA32 + credithistoryA33 +
## purposeA40 + purposeA41 + purposeA410 + purposeA45 + purposeA46 +
## purposeA48 + creditamount + savingsA61 + savingsA62 + savingsA63 +
## employmentlengthA71 + employmentlengthA72 + employmentlengthA73 +
## employmentlengthA74 + installmentrate + marriagesexA91 +
## marriagesexA92 + marriagesexA93 + otherdebtorsA101 + otherdebtorsA102 +
## propertyA121 + propertyA122 + propertyA123 + age + otherinstallmentplansA141 +
## otherinstallmentplansA142 + housingA151 + housingA152 + existingcredits +
## jobA171 + peopleliableno + telephone + foreignworker
##
## Df Deviance AIC
## - marriagesexA92 1 151.38 1033.9
## - employmentlengthA73 1 151.38 1033.9
## - housingA152 1 151.40 1034.1
## - jobA171 1 151.41 1034.1
## - peopleliableno 1 151.42 1034.2
## - employmentlengthA71 1 151.42 1034.2
## - purposeA410 1 151.43 1034.2
## - savingsA63 1 151.43 1034.2
## - marriagesexA93 1 151.44 1034.3
## - purposeA48 1 151.46 1034.4
## - purposeA45 1 151.48 1034.6
## - employmentlengthA72 1 151.50 1034.7
## - marriagesexA91 1 151.53 1034.9
## - otherinstallmentplansA142 1 151.56 1035.1
## - age 1 151.57 1035.2
## - propertyA122 1 151.60 1035.3
## <none> 151.32 1035.5
## - credithistoryA33 1 151.63 1035.6
## - telephone 1 151.64 1035.7
## - existingcheckingstatusA13 1 151.65 1035.7
## - existingcredits 1 151.66 1035.7
## - propertyA123 1 151.71 1036.1
## - employmentlengthA74 1 151.72 1036.2
## - savingsA62 1 151.74 1036.3
## - housingA151 1 151.77 1036.5
## - creditamount 1 151.83 1036.9
## - foreignworker 1 151.96 1037.8
## - propertyA121 1 151.97 1037.8
## - otherinstallmentplansA141 1 152.01 1038.1
## - purposeA41 1 152.18 1039.2
## - purposeA46 1 152.28 1039.8
## - otherdebtorsA101 1 152.61 1042.0
## - otherdebtorsA102 1 152.61 1042.0
## - installmentrate 1 152.80 1043.3
## - credithistoryA32 1 153.01 1044.6
## - duration 1 153.32 1046.7
## - credithistoryA30 1 153.63 1048.7
## - credithistoryA31 1 153.76 1049.5
## - savingsA61 1 153.80 1049.8
## - purposeA40 1 153.86 1050.2
## - existingcheckingstatusA12 1 156.62 1067.9
## - existingcheckingstatusA11 1 161.19 1096.7
##
## Step: AIC=1033.91
## classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
## existingcheckingstatusA13 + duration + credithistoryA30 +
## credithistoryA31 + credithistoryA32 + credithistoryA33 +
## purposeA40 + purposeA41 + purposeA410 + purposeA45 + purposeA46 +
## purposeA48 + creditamount + savingsA61 + savingsA62 + savingsA63 +
## employmentlengthA71 + employmentlengthA72 + employmentlengthA73 +
## employmentlengthA74 + installmentrate + marriagesexA91 +
## marriagesexA93 + otherdebtorsA101 + otherdebtorsA102 + propertyA121 +
## propertyA122 + propertyA123 + age + otherinstallmentplansA141 +
## otherinstallmentplansA142 + housingA151 + housingA152 + existingcredits +
## jobA171 + peopleliableno + telephone + foreignworker
##
## Df Deviance AIC
## - employmentlengthA73 1 151.44 1032.3
## - jobA171 1 151.46 1032.5
## - housingA152 1 151.47 1032.5
## - purposeA410 1 151.48 1032.6
## - savingsA63 1 151.48 1032.6
## - peopleliableno 1 151.48 1032.6
## - employmentlengthA71 1 151.48 1032.6
## - purposeA48 1 151.53 1032.9
## - marriagesexA91 1 151.53 1032.9
## - purposeA45 1 151.54 1032.9
## - employmentlengthA72 1 151.56 1033.1
## - age 1 151.61 1033.5
## - otherinstallmentplansA142 1 151.62 1033.5
## - propertyA122 1 151.66 1033.8
## - credithistoryA33 1 151.68 1033.9
## <none> 151.38 1033.9
## - existingcredits 1 151.71 1034.1
## - existingcheckingstatusA13 1 151.71 1034.1
## - telephone 1 151.72 1034.1
## - propertyA123 1 151.78 1034.5
## - employmentlengthA74 1 151.78 1034.6
## - savingsA62 1 151.79 1034.6
## - housingA151 1 151.84 1034.9
## - creditamount 1 151.91 1035.4
## - propertyA121 1 152.05 1036.3
## - foreignworker 1 152.05 1036.3
## - otherinstallmentplansA141 1 152.06 1036.4
## - marriagesexA93 1 152.12 1036.8
## - purposeA41 1 152.25 1037.6
## - purposeA46 1 152.37 1038.4
## - otherdebtorsA101 1 152.68 1040.4
## - otherdebtorsA102 1 152.68 1040.5
## - installmentrate 1 152.85 1041.6
## - credithistoryA32 1 153.05 1042.9
## - duration 1 153.38 1045.0
## - credithistoryA30 1 153.70 1047.1
## - credithistoryA31 1 153.84 1048.0
## - savingsA61 1 153.84 1048.0
## - purposeA40 1 153.93 1048.6
## - existingcheckingstatusA12 1 156.63 1066.0
## - existingcheckingstatusA11 1 161.25 1095.0
##
## Step: AIC=1032.33
## classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
## existingcheckingstatusA13 + duration + credithistoryA30 +
## credithistoryA31 + credithistoryA32 + credithistoryA33 +
## purposeA40 + purposeA41 + purposeA410 + purposeA45 + purposeA46 +
## purposeA48 + creditamount + savingsA61 + savingsA62 + savingsA63 +
## employmentlengthA71 + employmentlengthA72 + employmentlengthA74 +
## installmentrate + marriagesexA91 + marriagesexA93 + otherdebtorsA101 +
## otherdebtorsA102 + propertyA121 + propertyA122 + propertyA123 +
## age + otherinstallmentplansA141 + otherinstallmentplansA142 +
## housingA151 + housingA152 + existingcredits + jobA171 + peopleliableno +
## telephone + foreignworker
##
## Df Deviance AIC
## - employmentlengthA71 1 151.51 1030.8
## - jobA171 1 151.53 1030.9
## - housingA152 1 151.53 1030.9
## - peopleliableno 1 151.54 1031.0
## - purposeA410 1 151.54 1031.0
## - savingsA63 1 151.55 1031.0
## - employmentlengthA72 1 151.56 1031.1
## - purposeA48 1 151.59 1031.3
## - marriagesexA91 1 151.59 1031.3
## - purposeA45 1 151.60 1031.4
## - otherinstallmentplansA142 1 151.69 1032.0
## - propertyA122 1 151.71 1032.1
## <none> 151.44 1032.3
## - existingcredits 1 151.76 1032.4
## - credithistoryA33 1 151.76 1032.5
## - existingcheckingstatusA13 1 151.77 1032.5
## - age 1 151.78 1032.5
## - telephone 1 151.79 1032.6
## - propertyA123 1 151.83 1032.9
## - savingsA62 1 151.86 1033.1
## - housingA151 1 151.89 1033.3
## - creditamount 1 151.98 1033.9
## - propertyA121 1 152.08 1034.5
## - foreignworker 1 152.10 1034.6
## - otherinstallmentplansA141 1 152.10 1034.7
## - marriagesexA93 1 152.21 1035.4
## - employmentlengthA74 1 152.27 1035.8
## - purposeA41 1 152.32 1036.1
## - purposeA46 1 152.46 1037.0
## - otherdebtorsA101 1 152.72 1038.7
## - otherdebtorsA102 1 152.72 1038.7
## - installmentrate 1 152.88 1039.8
## - credithistoryA32 1 153.15 1041.5
## - duration 1 153.43 1043.4
## - credithistoryA30 1 153.82 1045.9
## - credithistoryA31 1 153.95 1046.8
## - savingsA61 1 153.96 1046.8
## - purposeA40 1 154.01 1047.2
## - existingcheckingstatusA12 1 156.68 1064.3
## - existingcheckingstatusA11 1 161.29 1093.3
##
## Step: AIC=1030.77
## classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
## existingcheckingstatusA13 + duration + credithistoryA30 +
## credithistoryA31 + credithistoryA32 + credithistoryA33 +
## purposeA40 + purposeA41 + purposeA410 + purposeA45 + purposeA46 +
## purposeA48 + creditamount + savingsA61 + savingsA62 + savingsA63 +
## employmentlengthA72 + employmentlengthA74 + installmentrate +
## marriagesexA91 + marriagesexA93 + otherdebtorsA101 + otherdebtorsA102 +
## propertyA121 + propertyA122 + propertyA123 + age + otherinstallmentplansA141 +
## otherinstallmentplansA142 + housingA151 + housingA152 + existingcredits +
## jobA171 + peopleliableno + telephone + foreignworker
##
## Df Deviance AIC
## - jobA171 1 151.55 1029.0
## - housingA152 1 151.59 1029.3
## - purposeA410 1 151.60 1029.4
## - peopleliableno 1 151.60 1029.4
## - employmentlengthA72 1 151.61 1029.4
## - savingsA63 1 151.61 1029.5
## - purposeA48 1 151.65 1029.7
## - marriagesexA91 1 151.66 1029.8
## - purposeA45 1 151.67 1029.9
## - propertyA122 1 151.77 1030.5
## - otherinstallmentplansA142 1 151.78 1030.5
## <none> 151.51 1030.8
## - existingcredits 1 151.82 1030.8
## - existingcheckingstatusA13 1 151.82 1030.8
## - age 1 151.83 1030.9
## - credithistoryA33 1 151.83 1030.9
## - telephone 1 151.84 1030.9
## - propertyA123 1 151.90 1031.3
## - savingsA62 1 151.95 1031.7
## - housingA151 1 151.95 1031.7
## - creditamount 1 152.08 1032.5
## - propertyA121 1 152.16 1033.1
## - otherinstallmentplansA141 1 152.16 1033.1
## - foreignworker 1 152.19 1033.2
## - marriagesexA93 1 152.30 1034.0
## - purposeA41 1 152.37 1034.4
## - employmentlengthA74 1 152.41 1034.7
## - purposeA46 1 152.50 1035.3
## - otherdebtorsA101 1 152.81 1037.3
## - otherdebtorsA102 1 152.83 1037.5
## - installmentrate 1 152.95 1038.2
## - credithistoryA32 1 153.21 1039.9
## - duration 1 153.45 1041.5
## - credithistoryA30 1 153.85 1044.1
## - credithistoryA31 1 154.02 1045.2
## - savingsA61 1 154.09 1045.6
## - purposeA40 1 154.09 1045.7
## - existingcheckingstatusA12 1 156.82 1063.2
## - existingcheckingstatusA11 1 161.45 1092.3
##
## Step: AIC=1029.02
## classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
## existingcheckingstatusA13 + duration + credithistoryA30 +
## credithistoryA31 + credithistoryA32 + credithistoryA33 +
## purposeA40 + purposeA41 + purposeA410 + purposeA45 + purposeA46 +
## purposeA48 + creditamount + savingsA61 + savingsA62 + savingsA63 +
## employmentlengthA72 + employmentlengthA74 + installmentrate +
## marriagesexA91 + marriagesexA93 + otherdebtorsA101 + otherdebtorsA102 +
## propertyA121 + propertyA122 + propertyA123 + age + otherinstallmentplansA141 +
## otherinstallmentplansA142 + housingA151 + housingA152 + existingcredits +
## peopleliableno + telephone + foreignworker
##
## Df Deviance AIC
## - housingA152 1 151.63 1027.6
## - purposeA410 1 151.64 1027.7
## - employmentlengthA72 1 151.64 1027.7
## - peopleliableno 1 151.64 1027.7
## - savingsA63 1 151.65 1027.7
## - purposeA48 1 151.69 1028.0
## - purposeA45 1 151.70 1028.0
## - marriagesexA91 1 151.71 1028.1
## - propertyA122 1 151.81 1028.7
## - otherinstallmentplansA142 1 151.82 1028.8
## - existingcheckingstatusA13 1 151.85 1029.0
## <none> 151.55 1029.0
## - existingcredits 1 151.85 1029.0
## - telephone 1 151.87 1029.2
## - credithistoryA33 1 151.88 1029.2
## - age 1 151.88 1029.2
## - propertyA123 1 151.93 1029.5
## - housingA151 1 151.99 1029.9
## - savingsA62 1 152.01 1030.0
## - creditamount 1 152.14 1030.9
## - propertyA121 1 152.19 1031.3
## - otherinstallmentplansA141 1 152.21 1031.4
## - foreignworker 1 152.23 1031.5
## - marriagesexA93 1 152.32 1032.1
## - purposeA41 1 152.41 1032.7
## - employmentlengthA74 1 152.43 1032.8
## - purposeA46 1 152.54 1033.6
## - otherdebtorsA101 1 152.83 1035.5
## - otherdebtorsA102 1 152.85 1035.6
## - installmentrate 1 153.05 1036.9
## - credithistoryA32 1 153.25 1038.2
## - duration 1 153.47 1039.7
## - credithistoryA30 1 153.90 1042.4
## - credithistoryA31 1 154.03 1043.2
## - purposeA40 1 154.10 1043.7
## - savingsA61 1 154.14 1044.0
## - existingcheckingstatusA12 1 156.82 1061.2
## - existingcheckingstatusA11 1 161.46 1090.4
##
## Step: AIC=1027.56
## classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
## existingcheckingstatusA13 + duration + credithistoryA30 +
## credithistoryA31 + credithistoryA32 + credithistoryA33 +
## purposeA40 + purposeA41 + purposeA410 + purposeA45 + purposeA46 +
## purposeA48 + creditamount + savingsA61 + savingsA62 + savingsA63 +
## employmentlengthA72 + employmentlengthA74 + installmentrate +
## marriagesexA91 + marriagesexA93 + otherdebtorsA101 + otherdebtorsA102 +
## propertyA121 + propertyA122 + propertyA123 + age + otherinstallmentplansA141 +
## otherinstallmentplansA142 + housingA151 + existingcredits +
## peopleliableno + telephone + foreignworker
##
## Df Deviance AIC
## - peopleliableno 1 151.71 1026.1
## - savingsA63 1 151.73 1026.2
## - purposeA410 1 151.73 1026.2
## - employmentlengthA72 1 151.74 1026.3
## - purposeA48 1 151.77 1026.5
## - purposeA45 1 151.78 1026.6
## - marriagesexA91 1 151.80 1026.7
## - propertyA122 1 151.81 1026.8
## - otherinstallmentplansA142 1 151.91 1027.4
## - existingcheckingstatusA13 1 151.92 1027.5
## - existingcredits 1 151.93 1027.5
## <none> 151.63 1027.6
## - telephone 1 151.95 1027.6
## - credithistoryA33 1 151.95 1027.7
## - propertyA123 1 151.98 1027.9
## - age 1 152.00 1028.0
## - savingsA62 1 152.09 1028.6
## - creditamount 1 152.23 1029.5
## - foreignworker 1 152.30 1030.0
## - otherinstallmentplansA141 1 152.30 1030.0
## - propertyA121 1 152.32 1030.1
## - housingA151 1 152.34 1030.2
## - marriagesexA93 1 152.39 1030.6
## - employmentlengthA74 1 152.49 1031.2
## - purposeA41 1 152.55 1031.6
## - purposeA46 1 152.60 1032.0
## - otherdebtorsA101 1 152.90 1033.9
## - otherdebtorsA102 1 152.96 1034.3
## - installmentrate 1 153.14 1035.5
## - credithistoryA32 1 153.30 1036.5
## - duration 1 153.53 1038.0
## - credithistoryA30 1 153.97 1040.9
## - credithistoryA31 1 154.09 1041.7
## - purposeA40 1 154.15 1042.1
## - savingsA61 1 154.19 1042.3
## - existingcheckingstatusA12 1 156.89 1059.7
## - existingcheckingstatusA11 1 161.47 1088.4
##
## Step: AIC=1026.12
## classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
## existingcheckingstatusA13 + duration + credithistoryA30 +
## credithistoryA31 + credithistoryA32 + credithistoryA33 +
## purposeA40 + purposeA41 + purposeA410 + purposeA45 + purposeA46 +
## purposeA48 + creditamount + savingsA61 + savingsA62 + savingsA63 +
## employmentlengthA72 + employmentlengthA74 + installmentrate +
## marriagesexA91 + marriagesexA93 + otherdebtorsA101 + otherdebtorsA102 +
## propertyA121 + propertyA122 + propertyA123 + age + otherinstallmentplansA141 +
## otherinstallmentplansA142 + housingA151 + existingcredits +
## telephone + foreignworker
##
## Df Deviance AIC
## - savingsA63 1 151.81 1024.8
## - purposeA410 1 151.81 1024.8
## - employmentlengthA72 1 151.83 1024.9
## - purposeA48 1 151.85 1025.0
## - purposeA45 1 151.88 1025.2
## - marriagesexA91 1 151.89 1025.2
## - propertyA122 1 151.91 1025.4
## - otherinstallmentplansA142 1 152.00 1026.0
## - existingcheckingstatusA13 1 152.00 1026.0
## <none> 151.71 1026.1
## - existingcredits 1 152.03 1026.2
## - telephone 1 152.04 1026.3
## - credithistoryA33 1 152.05 1026.3
## - age 1 152.08 1026.5
## - propertyA123 1 152.08 1026.5
## - savingsA62 1 152.18 1027.2
## - creditamount 1 152.30 1028.0
## - foreignworker 1 152.37 1028.4
## - marriagesexA93 1 152.40 1028.6
## - propertyA121 1 152.41 1028.7
## - otherinstallmentplansA141 1 152.41 1028.7
## - housingA151 1 152.42 1028.8
## - employmentlengthA74 1 152.57 1029.7
## - purposeA41 1 152.61 1030.0
## - purposeA46 1 152.73 1030.8
## - otherdebtorsA101 1 152.97 1032.3
## - otherdebtorsA102 1 153.02 1032.7
## - installmentrate 1 153.16 1033.6
## - credithistoryA32 1 153.41 1035.2
## - duration 1 153.60 1036.5
## - credithistoryA30 1 154.07 1039.5
## - credithistoryA31 1 154.25 1040.7
## - savingsA61 1 154.26 1040.7
## - purposeA40 1 154.35 1041.3
## - existingcheckingstatusA12 1 156.93 1057.9
## - existingcheckingstatusA11 1 161.64 1087.5
##
## Step: AIC=1024.75
## classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
## existingcheckingstatusA13 + duration + credithistoryA30 +
## credithistoryA31 + credithistoryA32 + credithistoryA33 +
## purposeA40 + purposeA41 + purposeA410 + purposeA45 + purposeA46 +
## purposeA48 + creditamount + savingsA61 + savingsA62 + employmentlengthA72 +
## employmentlengthA74 + installmentrate + marriagesexA91 +
## marriagesexA93 + otherdebtorsA101 + otherdebtorsA102 + propertyA121 +
## propertyA122 + propertyA123 + age + otherinstallmentplansA141 +
## otherinstallmentplansA142 + housingA151 + existingcredits +
## telephone + foreignworker
##
## Df Deviance AIC
## - purposeA410 1 151.91 1023.4
## - employmentlengthA72 1 151.92 1023.5
## - purposeA48 1 151.94 1023.6
## - purposeA45 1 151.98 1023.9
## - marriagesexA91 1 151.99 1023.9
## - propertyA122 1 152.01 1024.1
## - existingcheckingstatusA13 1 152.08 1024.5
## - otherinstallmentplansA142 1 152.09 1024.6
## - existingcredits 1 152.11 1024.7
## <none> 151.81 1024.8
## - telephone 1 152.14 1024.9
## - credithistoryA33 1 152.15 1025.0
## - age 1 152.17 1025.1
## - propertyA123 1 152.17 1025.2
## - savingsA62 1 152.19 1025.3
## - creditamount 1 152.36 1026.4
## - foreignworker 1 152.48 1027.2
## - marriagesexA93 1 152.48 1027.2
## - otherinstallmentplansA141 1 152.51 1027.3
## - propertyA121 1 152.53 1027.5
## - housingA151 1 152.53 1027.5
## - employmentlengthA74 1 152.68 1028.5
## - purposeA41 1 152.75 1028.9
## - purposeA46 1 152.80 1029.3
## - otherdebtorsA101 1 153.07 1031.0
## - otherdebtorsA102 1 153.10 1031.2
## - installmentrate 1 153.21 1031.9
## - credithistoryA32 1 153.50 1033.8
## - duration 1 153.72 1035.3
## - credithistoryA30 1 154.18 1038.2
## - savingsA61 1 154.35 1039.4
## - credithistoryA31 1 154.37 1039.5
## - purposeA40 1 154.42 1039.8
## - existingcheckingstatusA12 1 156.97 1056.2
## - existingcheckingstatusA11 1 161.69 1085.8
##
## Step: AIC=1023.43
## classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
## existingcheckingstatusA13 + duration + credithistoryA30 +
## credithistoryA31 + credithistoryA32 + credithistoryA33 +
## purposeA40 + purposeA41 + purposeA45 + purposeA46 + purposeA48 +
## creditamount + savingsA61 + savingsA62 + employmentlengthA72 +
## employmentlengthA74 + installmentrate + marriagesexA91 +
## marriagesexA93 + otherdebtorsA101 + otherdebtorsA102 + propertyA121 +
## propertyA122 + propertyA123 + age + otherinstallmentplansA141 +
## otherinstallmentplansA142 + housingA151 + existingcredits +
## telephone + foreignworker
##
## Df Deviance AIC
## - employmentlengthA72 1 152.03 1022.2
## - purposeA48 1 152.05 1022.3
## - purposeA45 1 152.09 1022.6
## - marriagesexA91 1 152.09 1022.6
## - propertyA122 1 152.11 1022.7
## - existingcheckingstatusA13 1 152.19 1023.2
## - otherinstallmentplansA142 1 152.19 1023.2
## - existingcredits 1 152.21 1023.4
## <none> 151.91 1023.4
## - credithistoryA33 1 152.24 1023.6
## - propertyA123 1 152.27 1023.8
## - telephone 1 152.28 1023.8
## - age 1 152.28 1023.9
## - savingsA62 1 152.29 1023.9
## - creditamount 1 152.44 1024.9
## - otherinstallmentplansA141 1 152.56 1025.7
## - marriagesexA93 1 152.57 1025.8
## - foreignworker 1 152.61 1026.0
## - propertyA121 1 152.62 1026.0
## - housingA151 1 152.66 1026.3
## - employmentlengthA74 1 152.75 1026.9
## - purposeA41 1 152.80 1027.3
## - purposeA46 1 152.94 1028.1
## - otherdebtorsA102 1 153.15 1029.5
## - otherdebtorsA101 1 153.18 1029.8
## - installmentrate 1 153.32 1030.7
## - credithistoryA32 1 153.60 1032.5
## - duration 1 153.82 1033.9
## - credithistoryA30 1 154.29 1036.9
## - savingsA61 1 154.44 1037.9
## - credithistoryA31 1 154.45 1038.0
## - purposeA40 1 154.64 1039.2
## - existingcheckingstatusA12 1 157.00 1054.3
## - existingcheckingstatusA11 1 161.72 1084.0
##
## Step: AIC=1022.22
## classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
## existingcheckingstatusA13 + duration + credithistoryA30 +
## credithistoryA31 + credithistoryA32 + credithistoryA33 +
## purposeA40 + purposeA41 + purposeA45 + purposeA46 + purposeA48 +
## creditamount + savingsA61 + savingsA62 + employmentlengthA74 +
## installmentrate + marriagesexA91 + marriagesexA93 + otherdebtorsA101 +
## otherdebtorsA102 + propertyA121 + propertyA122 + propertyA123 +
## age + otherinstallmentplansA141 + otherinstallmentplansA142 +
## housingA151 + existingcredits + telephone + foreignworker
##
## Df Deviance AIC
## - purposeA48 1 152.17 1021.1
## - marriagesexA91 1 152.21 1021.4
## - propertyA122 1 152.21 1021.4
## - purposeA45 1 152.21 1021.4
## - otherinstallmentplansA142 1 152.32 1022.1
## - existingcredits 1 152.32 1022.1
## - existingcheckingstatusA13 1 152.33 1022.1
## <none> 152.03 1022.2
## - credithistoryA33 1 152.37 1022.4
## - propertyA123 1 152.40 1022.6
## - telephone 1 152.40 1022.7
## - savingsA62 1 152.41 1022.7
## - age 1 152.49 1023.2
## - creditamount 1 152.57 1023.8
## - otherinstallmentplansA141 1 152.69 1024.5
## - foreignworker 1 152.71 1024.7
## - propertyA121 1 152.73 1024.8
## - marriagesexA93 1 152.80 1025.2
## - housingA151 1 152.80 1025.3
## - purposeA41 1 152.93 1026.1
## - employmentlengthA74 1 153.05 1026.9
## - purposeA46 1 153.06 1027.0
## - otherdebtorsA102 1 153.29 1028.5
## - otherdebtorsA101 1 153.35 1028.9
## - installmentrate 1 153.47 1029.6
## - credithistoryA32 1 153.76 1031.5
## - duration 1 153.91 1032.5
## - credithistoryA30 1 154.43 1035.9
## - credithistoryA31 1 154.60 1037.0
## - savingsA61 1 154.66 1037.4
## - purposeA40 1 154.78 1038.1
## - existingcheckingstatusA12 1 157.25 1054.0
## - existingcheckingstatusA11 1 161.85 1082.8
##
## Step: AIC=1021.14
## classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
## existingcheckingstatusA13 + duration + credithistoryA30 +
## credithistoryA31 + credithistoryA32 + credithistoryA33 +
## purposeA40 + purposeA41 + purposeA45 + purposeA46 + creditamount +
## savingsA61 + savingsA62 + employmentlengthA74 + installmentrate +
## marriagesexA91 + marriagesexA93 + otherdebtorsA101 + otherdebtorsA102 +
## propertyA121 + propertyA122 + propertyA123 + age + otherinstallmentplansA141 +
## otherinstallmentplansA142 + housingA151 + existingcredits +
## telephone + foreignworker
##
## Df Deviance AIC
## - marriagesexA91 1 152.35 1020.3
## - purposeA45 1 152.36 1020.4
## - propertyA122 1 152.36 1020.4
## - existingcredits 1 152.45 1021.0
## - otherinstallmentplansA142 1 152.47 1021.1
## <none> 152.17 1021.1
## - existingcheckingstatusA13 1 152.48 1021.1
## - credithistoryA33 1 152.50 1021.3
## - propertyA123 1 152.54 1021.5
## - telephone 1 152.54 1021.6
## - savingsA62 1 152.58 1021.8
## - age 1 152.63 1022.2
## - creditamount 1 152.72 1022.7
## - foreignworker 1 152.84 1023.5
## - otherinstallmentplansA141 1 152.85 1023.6
## - propertyA121 1 152.87 1023.7
## - marriagesexA93 1 152.91 1024.0
## - housingA151 1 152.97 1024.3
## - purposeA41 1 153.06 1024.9
## - employmentlengthA74 1 153.23 1026.1
## - purposeA46 1 153.23 1026.1
## - otherdebtorsA102 1 153.42 1027.3
## - otherdebtorsA101 1 153.47 1027.6
## - installmentrate 1 153.58 1028.3
## - credithistoryA32 1 153.87 1030.2
## - duration 1 154.11 1031.8
## - credithistoryA30 1 154.51 1034.4
## - credithistoryA31 1 154.65 1035.3
## - savingsA61 1 154.86 1036.7
## - purposeA40 1 155.01 1037.6
## - existingcheckingstatusA12 1 157.37 1052.7
## - existingcheckingstatusA11 1 162.01 1081.8
##
## Step: AIC=1020.34
## classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
## existingcheckingstatusA13 + duration + credithistoryA30 +
## credithistoryA31 + credithistoryA32 + credithistoryA33 +
## purposeA40 + purposeA41 + purposeA45 + purposeA46 + creditamount +
## savingsA61 + savingsA62 + employmentlengthA74 + installmentrate +
## marriagesexA93 + otherdebtorsA101 + otherdebtorsA102 + propertyA121 +
## propertyA122 + propertyA123 + age + otherinstallmentplansA141 +
## otherinstallmentplansA142 + housingA151 + existingcredits +
## telephone + foreignworker
##
## Df Deviance AIC
## - propertyA122 1 152.54 1019.6
## - purposeA45 1 152.55 1019.6
## - existingcredits 1 152.64 1020.2
## - otherinstallmentplansA142 1 152.64 1020.2
## <none> 152.35 1020.3
## - existingcheckingstatusA13 1 152.69 1020.5
## - credithistoryA33 1 152.69 1020.6
## - propertyA123 1 152.70 1020.6
## - telephone 1 152.72 1020.7
## - age 1 152.76 1021.0
## - savingsA62 1 152.77 1021.1
## - creditamount 1 152.94 1022.1
## - otherinstallmentplansA141 1 153.01 1022.6
## - propertyA121 1 153.03 1022.7
## - foreignworker 1 153.03 1022.8
## - housingA151 1 153.09 1023.2
## - purposeA41 1 153.27 1024.3
## - purposeA46 1 153.38 1025.0
## - marriagesexA93 1 153.38 1025.0
## - employmentlengthA74 1 153.45 1025.5
## - otherdebtorsA102 1 153.60 1026.5
## - otherdebtorsA101 1 153.68 1027.0
## - installmentrate 1 153.71 1027.2
## - credithistoryA32 1 154.05 1029.4
## - duration 1 154.31 1031.1
## - credithistoryA30 1 154.68 1033.5
## - credithistoryA31 1 154.84 1034.5
## - savingsA61 1 155.02 1035.7
## - purposeA40 1 155.15 1036.5
## - existingcheckingstatusA12 1 157.59 1052.1
## - existingcheckingstatusA11 1 162.44 1082.4
##
## Step: AIC=1019.56
## classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
## existingcheckingstatusA13 + duration + credithistoryA30 +
## credithistoryA31 + credithistoryA32 + credithistoryA33 +
## purposeA40 + purposeA41 + purposeA45 + purposeA46 + creditamount +
## savingsA61 + savingsA62 + employmentlengthA74 + installmentrate +
## marriagesexA93 + otherdebtorsA101 + otherdebtorsA102 + propertyA121 +
## propertyA123 + age + otherinstallmentplansA141 + otherinstallmentplansA142 +
## housingA151 + existingcredits + telephone + foreignworker
##
## Df Deviance AIC
## - propertyA123 1 152.71 1018.6
## - purposeA45 1 152.76 1019.0
## - otherinstallmentplansA142 1 152.80 1019.3
## - existingcredits 1 152.81 1019.3
## <none> 152.54 1019.6
## - credithistoryA33 1 152.87 1019.7
## - telephone 1 152.88 1019.8
## - age 1 152.89 1019.8
## - existingcheckingstatusA13 1 152.90 1019.9
## - savingsA62 1 152.98 1020.4
## - propertyA121 1 153.05 1020.9
## - creditamount 1 153.18 1021.8
## - otherinstallmentplansA141 1 153.24 1022.1
## - foreignworker 1 153.25 1022.2
## - housingA151 1 153.27 1022.3
## - purposeA41 1 153.39 1023.1
## - marriagesexA93 1 153.51 1023.9
## - purposeA46 1 153.69 1025.1
## - employmentlengthA74 1 153.69 1025.1
## - otherdebtorsA102 1 153.89 1026.4
## - otherdebtorsA101 1 153.95 1026.7
## - installmentrate 1 153.96 1026.8
## - credithistoryA32 1 154.24 1028.6
## - duration 1 154.59 1030.9
## - credithistoryA30 1 154.87 1032.7
## - credithistoryA31 1 155.13 1034.4
## - savingsA61 1 155.29 1035.4
## - purposeA40 1 155.45 1036.5
## - existingcheckingstatusA12 1 157.88 1051.9
## - existingcheckingstatusA11 1 162.63 1081.6
##
## Step: AIC=1018.63
## classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
## existingcheckingstatusA13 + duration + credithistoryA30 +
## credithistoryA31 + credithistoryA32 + credithistoryA33 +
## purposeA40 + purposeA41 + purposeA45 + purposeA46 + creditamount +
## savingsA61 + savingsA62 + employmentlengthA74 + installmentrate +
## marriagesexA93 + otherdebtorsA101 + otherdebtorsA102 + propertyA121 +
## age + otherinstallmentplansA141 + otherinstallmentplansA142 +
## housingA151 + existingcredits + telephone + foreignworker
##
## Df Deviance AIC
## - purposeA45 1 152.93 1018.1
## - existingcredits 1 152.95 1018.2
## - otherinstallmentplansA142 1 152.96 1018.3
## - age 1 153.00 1018.6
## <none> 152.71 1018.6
## - credithistoryA33 1 153.04 1018.8
## - telephone 1 153.05 1018.9
## - existingcheckingstatusA13 1 153.06 1019.0
## - propertyA121 1 153.07 1019.0
## - savingsA62 1 153.14 1019.5
## - creditamount 1 153.38 1021.1
## - foreignworker 1 153.38 1021.1
## - otherinstallmentplansA141 1 153.42 1021.3
## - housingA151 1 153.43 1021.4
## - purposeA41 1 153.56 1022.2
## - marriagesexA93 1 153.64 1022.8
## - employmentlengthA74 1 153.91 1024.5
## - purposeA46 1 153.94 1024.7
## - otherdebtorsA102 1 154.04 1025.3
## - otherdebtorsA101 1 154.06 1025.5
## - installmentrate 1 154.14 1026.0
## - credithistoryA32 1 154.36 1027.4
## - duration 1 154.70 1029.6
## - credithistoryA30 1 155.01 1031.7
## - credithistoryA31 1 155.29 1033.4
## - savingsA61 1 155.40 1034.1
## - purposeA40 1 155.65 1035.7
## - existingcheckingstatusA12 1 158.12 1051.5
## - existingcheckingstatusA11 1 163.23 1083.3
##
## Step: AIC=1018.11
## classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
## existingcheckingstatusA13 + duration + credithistoryA30 +
## credithistoryA31 + credithistoryA32 + credithistoryA33 +
## purposeA40 + purposeA41 + purposeA46 + creditamount + savingsA61 +
## savingsA62 + employmentlengthA74 + installmentrate + marriagesexA93 +
## otherdebtorsA101 + otherdebtorsA102 + propertyA121 + age +
## otherinstallmentplansA141 + otherinstallmentplansA142 + housingA151 +
## existingcredits + telephone + foreignworker
##
## Df Deviance AIC
## - otherinstallmentplansA142 1 153.18 1017.7
## - age 1 153.19 1017.8
## - existingcredits 1 153.21 1017.9
## <none> 152.93 1018.1
## - existingcheckingstatusA13 1 153.27 1018.3
## - propertyA121 1 153.28 1018.4
## - credithistoryA33 1 153.28 1018.4
## - telephone 1 153.31 1018.6
## - savingsA62 1 153.38 1019.0
## - otherinstallmentplansA141 1 153.59 1020.4
## - foreignworker 1 153.60 1020.5
## - creditamount 1 153.62 1020.6
## - housingA151 1 153.64 1020.7
## - purposeA41 1 153.83 1022.0
## - marriagesexA93 1 153.88 1022.3
## - purposeA46 1 154.12 1023.9
## - employmentlengthA74 1 154.14 1024.0
## - otherdebtorsA102 1 154.26 1024.7
## - otherdebtorsA101 1 154.27 1024.8
## - installmentrate 1 154.40 1025.7
## - credithistoryA32 1 154.64 1027.2
## - duration 1 154.90 1028.9
## - credithistoryA30 1 155.30 1031.5
## - credithistoryA31 1 155.52 1032.9
## - savingsA61 1 155.64 1033.7
## - purposeA40 1 155.73 1034.3
## - existingcheckingstatusA12 1 158.44 1051.5
## - existingcheckingstatusA11 1 163.46 1082.7
##
## Step: AIC=1017.71
## classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
## existingcheckingstatusA13 + duration + credithistoryA30 +
## credithistoryA31 + credithistoryA32 + credithistoryA33 +
## purposeA40 + purposeA41 + purposeA46 + creditamount + savingsA61 +
## savingsA62 + employmentlengthA74 + installmentrate + marriagesexA93 +
## otherdebtorsA101 + otherdebtorsA102 + propertyA121 + age +
## otherinstallmentplansA141 + housingA151 + existingcredits +
## telephone + foreignworker
##
## Df Deviance AIC
## - age 1 153.46 1017.6
## - existingcredits 1 153.47 1017.6
## <none> 153.18 1017.7
## - propertyA121 1 153.51 1017.9
## - existingcheckingstatusA13 1 153.53 1018.0
## - telephone 1 153.54 1018.1
## - credithistoryA33 1 153.59 1018.4
## - savingsA62 1 153.61 1018.6
## - otherinstallmentplansA141 1 153.75 1019.5
## - housingA151 1 153.82 1019.9
## - foreignworker 1 153.85 1020.1
## - creditamount 1 153.87 1020.2
## - marriagesexA93 1 154.08 1021.6
## - purposeA41 1 154.09 1021.7
## - purposeA46 1 154.35 1023.3
## - employmentlengthA74 1 154.43 1023.9
## - otherdebtorsA102 1 154.50 1024.3
## - otherdebtorsA101 1 154.54 1024.6
## - installmentrate 1 154.69 1025.5
## - credithistoryA32 1 154.91 1027.0
## - duration 1 155.18 1028.7
## - credithistoryA30 1 155.58 1031.3
## - purposeA40 1 155.89 1033.3
## - savingsA61 1 155.91 1033.4
## - credithistoryA31 1 156.09 1034.6
## - existingcheckingstatusA12 1 158.64 1050.8
## - existingcheckingstatusA11 1 163.74 1082.4
##
## Step: AIC=1017.56
## classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
## existingcheckingstatusA13 + duration + credithistoryA30 +
## credithistoryA31 + credithistoryA32 + credithistoryA33 +
## purposeA40 + purposeA41 + purposeA46 + creditamount + savingsA61 +
## savingsA62 + employmentlengthA74 + installmentrate + marriagesexA93 +
## otherdebtorsA101 + otherdebtorsA102 + propertyA121 + otherinstallmentplansA141 +
## housingA151 + existingcredits + telephone + foreignworker
##
## Df Deviance AIC
## - existingcredits 1 153.75 1017.4
## <none> 153.46 1017.6
## - existingcheckingstatusA13 1 153.79 1017.7
## - propertyA121 1 153.83 1018.0
## - credithistoryA33 1 153.90 1018.4
## - telephone 1 153.91 1018.5
## - savingsA62 1 153.95 1018.8
## - otherinstallmentplansA141 1 154.00 1019.1
## - foreignworker 1 154.12 1019.9
## - creditamount 1 154.12 1019.9
## - housingA151 1 154.28 1020.9
## - purposeA41 1 154.43 1021.8
## - marriagesexA93 1 154.53 1022.5
## - purposeA46 1 154.55 1022.6
## - employmentlengthA74 1 154.61 1023.0
## - otherdebtorsA102 1 154.79 1024.2
## - otherdebtorsA101 1 154.83 1024.5
## - installmentrate 1 154.91 1025.0
## - credithistoryA32 1 155.36 1027.9
## - duration 1 155.59 1029.4
## - credithistoryA30 1 155.92 1031.5
## - purposeA40 1 156.05 1032.3
## - savingsA61 1 156.34 1034.1
## - credithistoryA31 1 156.43 1034.7
## - existingcheckingstatusA12 1 159.02 1051.2
## - existingcheckingstatusA11 1 163.96 1081.8
##
## Step: AIC=1017.43
## classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
## existingcheckingstatusA13 + duration + credithistoryA30 +
## credithistoryA31 + credithistoryA32 + credithistoryA33 +
## purposeA40 + purposeA41 + purposeA46 + creditamount + savingsA61 +
## savingsA62 + employmentlengthA74 + installmentrate + marriagesexA93 +
## otherdebtorsA101 + otherdebtorsA102 + propertyA121 + otherinstallmentplansA141 +
## housingA151 + telephone + foreignworker
##
## Df Deviance AIC
## - existingcheckingstatusA13 1 154.05 1017.4
## <none> 153.75 1017.4
## - propertyA121 1 154.12 1017.8
## - credithistoryA33 1 154.14 1018.0
## - telephone 1 154.20 1018.4
## - savingsA62 1 154.25 1018.7
## - otherinstallmentplansA141 1 154.33 1019.2
## - creditamount 1 154.41 1019.7
## - foreignworker 1 154.43 1019.9
## - housingA151 1 154.57 1020.8
## - purposeA41 1 154.72 1021.8
## - marriagesexA93 1 154.79 1022.2
## - purposeA46 1 154.79 1022.2
## - employmentlengthA74 1 154.85 1022.6
## - otherdebtorsA101 1 155.08 1024.0
## - otherdebtorsA102 1 155.08 1024.0
## - installmentrate 1 155.23 1025.0
## - credithistoryA32 1 155.50 1026.8
## - duration 1 155.85 1029.0
## - credithistoryA30 1 156.15 1031.0
## - purposeA40 1 156.37 1032.4
## - credithistoryA31 1 156.43 1032.7
## - savingsA61 1 156.67 1034.2
## - existingcheckingstatusA12 1 159.19 1050.2
## - existingcheckingstatusA11 1 164.23 1081.4
##
## Step: AIC=1017.41
## classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
## duration + credithistoryA30 + credithistoryA31 + credithistoryA32 +
## credithistoryA33 + purposeA40 + purposeA41 + purposeA46 +
## creditamount + savingsA61 + savingsA62 + employmentlengthA74 +
## installmentrate + marriagesexA93 + otherdebtorsA101 + otherdebtorsA102 +
## propertyA121 + otherinstallmentplansA141 + housingA151 +
## telephone + foreignworker
##
## Df Deviance AIC
## <none> 154.05 1017.4
## - propertyA121 1 154.43 1017.9
## - credithistoryA33 1 154.44 1017.9
## - telephone 1 154.50 1018.4
## - savingsA62 1 154.57 1018.8
## - otherinstallmentplansA141 1 154.64 1019.2
## - creditamount 1 154.65 1019.3
## - foreignworker 1 154.71 1019.7
## - housingA151 1 154.86 1020.6
## - purposeA41 1 155.08 1022.1
## - marriagesexA93 1 155.09 1022.1
## - purposeA46 1 155.12 1022.3
## - employmentlengthA74 1 155.21 1022.9
## - otherdebtorsA102 1 155.40 1024.1
## - otherdebtorsA101 1 155.41 1024.2
## - installmentrate 1 155.44 1024.4
## - credithistoryA32 1 155.89 1027.3
## - duration 1 156.22 1029.4
## - credithistoryA30 1 156.51 1031.2
## - purposeA40 1 156.71 1032.5
## - credithistoryA31 1 156.83 1033.3
## - savingsA61 1 157.14 1035.2
## - existingcheckingstatusA12 1 159.19 1048.2
## - existingcheckingstatusA11 1 164.31 1079.8step##
## Call: glm(formula = classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
## duration + credithistoryA30 + credithistoryA31 + credithistoryA32 +
## credithistoryA33 + purposeA40 + purposeA41 + purposeA46 +
## creditamount + savingsA61 + savingsA62 + employmentlengthA74 +
## installmentrate + marriagesexA93 + otherdebtorsA101 + otherdebtorsA102 +
## propertyA121 + otherinstallmentplansA141 + housingA151 +
## telephone + foreignworker, data = creditDataNumeric)
##
## Coefficients:
## (Intercept) existingcheckingstatusA11
## -4.072e-01 2.618e-01
## existingcheckingstatusA12 duration
## 1.824e-01 5.840e-03
## credithistoryA30 credithistoryA31
## 2.748e-01 2.682e-01
## credithistoryA32 credithistoryA33
## 1.014e-01 7.750e-02
## purposeA40 purposeA41
## 1.286e-01 -1.157e-01
## purposeA46 creditamount
## 1.540e-01 1.648e-05
## savingsA61 savingsA62
## 1.310e-01 8.410e-02
## employmentlengthA74 installmentrate
## -9.076e-02 3.866e-02
## marriagesexA93 otherdebtorsA101
## -6.951e-02 1.738e-01
## otherdebtorsA102 propertyA121
## 2.474e-01 -4.673e-02
## otherinstallmentplansA141 housingA151
## 7.261e-02 7.745e-02
## telephone foreignworker
## -4.609e-02 -1.418e-01
##
## Degrees of Freedom: 999 Total (i.e. Null); 976 Residual
## Null Deviance: 210
## Residual Deviance: 154.1 AIC: 1017model2 <- glm(formula = classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
duration + credithistoryA30 + credithistoryA31 + credithistoryA32 +
credithistoryA33 + purposeA40 + purposeA41 + purposeA46 +
creditamount + savingsA61 + savingsA62 + employmentlengthA74 +
installmentrate + marriagesexA93 + otherdebtorsA101 + otherdebtorsA102 +
propertyA121 + otherinstallmentplansA141 + housingA151 +
telephone + foreignworker, data = creditDataNumeric)
summary(model2)##
## Call:
## glm(formula = classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
## duration + credithistoryA30 + credithistoryA31 + credithistoryA32 +
## credithistoryA33 + purposeA40 + purposeA41 + purposeA46 +
## creditamount + savingsA61 + savingsA62 + employmentlengthA74 +
## installmentrate + marriagesexA93 + otherdebtorsA101 + otherdebtorsA102 +
## propertyA121 + otherinstallmentplansA141 + housingA151 +
## telephone + foreignworker, data = creditDataNumeric)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.88971 -0.28321 -0.09468 0.31168 1.12452
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -4.072e-01 8.644e-02 -4.711 2.82e-06 ***
## existingcheckingstatusA11 2.618e-01 3.248e-02 8.060 2.22e-15 ***
## existingcheckingstatusA12 1.824e-01 3.196e-02 5.705 1.54e-08 ***
## duration 5.840e-03 1.575e-03 3.709 0.00022 ***
## credithistoryA30 2.748e-01 6.966e-02 3.945 8.57e-05 ***
## credithistoryA31 2.682e-01 6.396e-02 4.194 2.99e-05 ***
## credithistoryA32 1.014e-01 2.967e-02 3.417 0.00066 ***
## credithistoryA33 7.750e-02 4.963e-02 1.562 0.11868
## purposeA40 1.286e-01 3.135e-02 4.102 4.44e-05 ***
## purposeA41 -1.157e-01 4.532e-02 -2.553 0.01082 *
## purposeA46 1.540e-01 5.925e-02 2.599 0.00948 **
## creditamount 1.648e-05 8.453e-06 1.949 0.05156 .
## savingsA61 1.310e-01 2.962e-02 4.422 1.09e-05 ***
## savingsA62 8.410e-02 4.649e-02 1.809 0.07076 .
## employmentlengthA74 -9.076e-02 3.352e-02 -2.708 0.00689 **
## installmentrate 3.866e-02 1.303e-02 2.966 0.00309 **
## marriagesexA93 -6.951e-02 2.709e-02 -2.566 0.01043 *
## otherdebtorsA101 1.738e-01 5.918e-02 2.937 0.00339 **
## otherdebtorsA102 2.474e-01 8.458e-02 2.925 0.00352 **
## propertyA121 -4.673e-02 3.018e-02 -1.548 0.12192
## otherinstallmentplansA141 7.261e-02 3.769e-02 1.927 0.05433 .
## housingA151 7.745e-02 3.423e-02 2.262 0.02389 *
## telephone -4.609e-02 2.720e-02 -1.695 0.09047 .
## foreignworker -1.418e-01 6.936e-02 -2.045 0.04113 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 0.1578397)
##
## Null deviance: 210.00 on 999 degrees of freedom
## Residual deviance: 154.05 on 976 degrees of freedom
## AIC: 1017.4
##
## Number of Fisher Scoring iterations: 2vif(model2)## existingcheckingstatusA11 existingcheckingstatusA12
## 1.329751 1.272792
## duration credithistoryA30
## 2.078904 1.180558
## credithistoryA31 credithistoryA32
## 1.207869 1.389213
## credithistoryA33 purposeA40
## 1.252176 1.116129
## purposeA41 purposeA46
## 1.202239 1.056385
## creditamount savingsA61
## 2.541479 1.330572
## savingsA62 employmentlengthA74
## 1.264991 1.022835
## installmentrate marriagesexA93
## 1.345320 1.151333
## otherdebtorsA101 otherdebtorsA102
## 1.871454 1.781867
## propertyA121 otherinstallmentplansA141
## 1.168594 1.077118
## housingA151 telephone
## 1.091093 1.128469
## foreignworker
## 1.086032## remove creditamount for having highest vif
model3 <- glm(formula = classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
duration + credithistoryA30 + credithistoryA31 + credithistoryA32 +
credithistoryA33 + purposeA40 + purposeA41 + purposeA46 +
savingsA61 + savingsA62 + employmentlengthA74 +
installmentrate + marriagesexA93 + otherdebtorsA101 + otherdebtorsA102 +
propertyA121 + otherinstallmentplansA141 + housingA151 +
telephone + foreignworker, data = creditDataNumeric)
summary(model3)##
## Call:
## glm(formula = classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
## duration + credithistoryA30 + credithistoryA31 + credithistoryA32 +
## credithistoryA33 + purposeA40 + purposeA41 + purposeA46 +
## savingsA61 + savingsA62 + employmentlengthA74 + installmentrate +
## marriagesexA93 + otherdebtorsA101 + otherdebtorsA102 + propertyA121 +
## otherinstallmentplansA141 + housingA151 + telephone + foreignworker,
## data = creditDataNumeric)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.85929 -0.28975 -0.09282 0.31562 1.09762
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -0.375079 0.084971 -4.414 1.13e-05 ***
## existingcheckingstatusA11 0.262585 0.032527 8.073 2.01e-15 ***
## existingcheckingstatusA12 0.186812 0.031927 5.851 6.65e-09 ***
## duration 0.007834 0.001199 6.533 1.03e-10 ***
## credithistoryA30 0.281376 0.069678 4.038 5.81e-05 ***
## credithistoryA31 0.266893 0.064051 4.167 3.36e-05 ***
## credithistoryA32 0.099879 0.029702 3.363 0.000802 ***
## credithistoryA33 0.081466 0.049654 1.641 0.101192
## purposeA40 0.131798 0.031352 4.204 2.87e-05 ***
## purposeA41 -0.096892 0.044342 -2.185 0.029119 *
## purposeA46 0.156835 0.059315 2.644 0.008322 **
## savingsA61 0.130331 0.029660 4.394 1.23e-05 ***
## savingsA62 0.080519 0.046518 1.731 0.083782 .
## employmentlengthA74 -0.090873 0.033563 -2.707 0.006898 **
## installmentrate 0.027037 0.011605 2.330 0.020020 *
## marriagesexA93 -0.061980 0.026847 -2.309 0.021174 *
## otherdebtorsA101 0.176024 0.059251 2.971 0.003043 **
## otherdebtorsA102 0.261861 0.084370 3.104 0.001966 **
## propertyA121 -0.052926 0.030057 -1.761 0.078576 .
## otherinstallmentplansA141 0.071280 0.037738 1.889 0.059213 .
## housingA151 0.078528 0.034277 2.291 0.022177 *
## telephone -0.035211 0.026657 -1.321 0.186853
## foreignworker -0.140778 0.069458 -2.027 0.042954 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 0.1582919)
##
## Null deviance: 210.00 on 999 degrees of freedom
## Residual deviance: 154.65 on 977 degrees of freedom
## AIC: 1019.3
##
## Number of Fisher Scoring iterations: 2vif(model3)## existingcheckingstatusA11 existingcheckingstatusA12
## 1.329546 1.266283
## duration credithistoryA30
## 1.201808 1.177771
## credithistoryA31 credithistoryA32
## 1.207727 1.388282
## credithistoryA33 purposeA40
## 1.250072 1.113074
## purposeA41 purposeA46
## 1.147637 1.055751
## savingsA61 savingsA62
## 1.330400 1.263020
## employmentlengthA74 installmentrate
## 1.022831 1.063719
## marriagesexA93 otherdebtorsA101
## 1.127891 1.870753
## otherdebtorsA102 propertyA121
## 1.768133 1.155612
## otherinstallmentplansA141 housingA151
## 1.076763 1.090807
## telephone foreignworker
## 1.080946 1.085966## all have low vif, so we can don't need to check vif anymore
## remove telephone as it has highest p-value
model4 <- glm(formula = classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
duration + credithistoryA30 + credithistoryA31 + credithistoryA32 +
credithistoryA33 + purposeA40 + purposeA41 + purposeA46 +
savingsA61 + savingsA62 + employmentlengthA74 +
installmentrate + marriagesexA93 + otherdebtorsA101 + otherdebtorsA102 +
propertyA121 + otherinstallmentplansA141 + housingA151 +
foreignworker, data = creditDataNumeric)
summary(model4)##
## Call:
## glm(formula = classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
## duration + credithistoryA30 + credithistoryA31 + credithistoryA32 +
## credithistoryA33 + purposeA40 + purposeA41 + purposeA46 +
## savingsA61 + savingsA62 + employmentlengthA74 + installmentrate +
## marriagesexA93 + otherdebtorsA101 + otherdebtorsA102 + propertyA121 +
## otherinstallmentplansA141 + housingA151 + foreignworker,
## data = creditDataNumeric)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.87356 -0.29191 -0.09868 0.31975 1.08000
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -0.386562 0.084557 -4.572 5.46e-06 ***
## existingcheckingstatusA11 0.264692 0.032500 8.144 1.16e-15 ***
## existingcheckingstatusA12 0.186791 0.031939 5.848 6.77e-09 ***
## duration 0.007664 0.001193 6.426 2.04e-10 ***
## credithistoryA30 0.283684 0.069682 4.071 5.06e-05 ***
## credithistoryA31 0.269502 0.064045 4.208 2.81e-05 ***
## credithistoryA32 0.101397 0.029691 3.415 0.000664 ***
## credithistoryA33 0.081725 0.049673 1.645 0.100238
## purposeA40 0.131224 0.031361 4.184 3.12e-05 ***
## purposeA41 -0.102257 0.044172 -2.315 0.020822 *
## purposeA46 0.156267 0.059336 2.634 0.008582 **
## savingsA61 0.132046 0.029643 4.455 9.37e-06 ***
## savingsA62 0.083917 0.046464 1.806 0.071216 .
## employmentlengthA74 -0.091795 0.033569 -2.735 0.006360 **
## installmentrate 0.027070 0.011609 2.332 0.019914 *
## marriagesexA93 -0.063194 0.026842 -2.354 0.018755 *
## otherdebtorsA101 0.173575 0.059244 2.930 0.003470 **
## otherdebtorsA102 0.259629 0.084385 3.077 0.002151 **
## propertyA121 -0.048784 0.029904 -1.631 0.103147
## otherinstallmentplansA141 0.070905 0.037751 1.878 0.060652 .
## housingA151 0.079464 0.034283 2.318 0.020662 *
## foreignworker -0.134124 0.069302 -1.935 0.053234 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 0.1584124)
##
## Null deviance: 210.00 on 999 degrees of freedom
## Residual deviance: 154.93 on 978 degrees of freedom
## AIC: 1019.1
##
## Number of Fisher Scoring iterations: 2## remove propertyA121 as it has highest p-value
model5 <- glm(formula = classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
duration + credithistoryA30 + credithistoryA31 + credithistoryA32 +
credithistoryA33 + purposeA40 + purposeA41 + purposeA46 +
savingsA61 + savingsA62 + employmentlengthA74 +
installmentrate + marriagesexA93 + otherdebtorsA101 + otherdebtorsA102 +
otherinstallmentplansA141 + housingA151 +
foreignworker, data = creditDataNumeric)
summary(model5)##
## Call:
## glm(formula = classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
## duration + credithistoryA30 + credithistoryA31 + credithistoryA32 +
## credithistoryA33 + purposeA40 + purposeA41 + purposeA46 +
## savingsA61 + savingsA62 + employmentlengthA74 + installmentrate +
## marriagesexA93 + otherdebtorsA101 + otherdebtorsA102 + otherinstallmentplansA141 +
## housingA151 + foreignworker, data = creditDataNumeric)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.87044 -0.28648 -0.09755 0.32841 1.09752
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -0.427395 0.080836 -5.287 1.53e-07 ***
## existingcheckingstatusA11 0.267445 0.032483 8.233 5.78e-16 ***
## existingcheckingstatusA12 0.187409 0.031964 5.863 6.21e-09 ***
## duration 0.008032 0.001172 6.853 1.28e-11 ***
## credithistoryA30 0.289589 0.069647 4.158 3.49e-05 ***
## credithistoryA31 0.270699 0.064095 4.223 2.63e-05 ***
## credithistoryA32 0.102249 0.029711 3.441 0.000603 ***
## credithistoryA33 0.082659 0.049712 1.663 0.096681 .
## purposeA40 0.132055 0.031384 4.208 2.82e-05 ***
## purposeA41 -0.094686 0.043965 -2.154 0.031511 *
## purposeA46 0.166761 0.059036 2.825 0.004828 **
## savingsA61 0.131306 0.029664 4.426 1.07e-05 ***
## savingsA62 0.087683 0.046446 1.888 0.059342 .
## employmentlengthA74 -0.090252 0.033584 -2.687 0.007324 **
## installmentrate 0.027324 0.011618 2.352 0.018877 *
## marriagesexA93 -0.060569 0.026816 -2.259 0.024125 *
## otherdebtorsA101 0.187991 0.058631 3.206 0.001388 **
## otherdebtorsA102 0.273405 0.084032 3.254 0.001179 **
## otherinstallmentplansA141 0.074739 0.037710 1.982 0.047767 *
## housingA151 0.079065 0.034311 2.304 0.021411 *
## foreignworker -0.142060 0.069189 -2.053 0.040318 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 0.1586812)
##
## Null deviance: 210.00 on 999 degrees of freedom
## Residual deviance: 155.35 on 979 degrees of freedom
## AIC: 1019.8
##
## Number of Fisher Scoring iterations: 2## remove credithistoryA33 as it has highest p-value
model6 <- glm(formula = classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
duration + credithistoryA30 + credithistoryA31 + credithistoryA32 +
purposeA40 + purposeA41 + purposeA46 +
savingsA61 + savingsA62 + employmentlengthA74 +
installmentrate + marriagesexA93 + otherdebtorsA101 + otherdebtorsA102 +
otherinstallmentplansA141 + housingA151 +
foreignworker, data = creditDataNumeric)
summary(model6)##
## Call:
## glm(formula = classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
## duration + credithistoryA30 + credithistoryA31 + credithistoryA32 +
## purposeA40 + purposeA41 + purposeA46 + savingsA61 + savingsA62 +
## employmentlengthA74 + installmentrate + marriagesexA93 +
## otherdebtorsA101 + otherdebtorsA102 + otherinstallmentplansA141 +
## housingA151 + foreignworker, data = creditDataNumeric)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.87240 -0.29227 -0.09837 0.32457 1.07965
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -0.418050 0.080713 -5.179 2.70e-07 ***
## existingcheckingstatusA11 0.267247 0.032512 8.220 6.41e-16 ***
## existingcheckingstatusA12 0.192382 0.031853 6.040 2.19e-09 ***
## duration 0.008287 0.001163 7.126 2.01e-12 ***
## credithistoryA30 0.266075 0.068258 3.898 0.000104 ***
## credithistoryA31 0.250844 0.063029 3.980 7.41e-05 ***
## credithistoryA32 0.083290 0.027461 3.033 0.002485 **
## purposeA40 0.129758 0.031381 4.135 3.86e-05 ***
## purposeA41 -0.099421 0.043912 -2.264 0.023788 *
## purposeA46 0.164031 0.059066 2.777 0.005590 **
## savingsA61 0.132215 0.029686 4.454 9.41e-06 ***
## savingsA62 0.092469 0.046399 1.993 0.046548 *
## employmentlengthA74 -0.091028 0.033611 -2.708 0.006881 **
## installmentrate 0.026378 0.011615 2.271 0.023356 *
## marriagesexA93 -0.058852 0.026821 -2.194 0.028449 *
## otherdebtorsA101 0.193805 0.058579 3.308 0.000972 ***
## otherdebtorsA102 0.273010 0.084108 3.246 0.001210 **
## otherinstallmentplansA141 0.075663 0.037740 2.005 0.045255 *
## housingA151 0.080179 0.034335 2.335 0.019735 *
## foreignworker -0.145796 0.069215 -2.106 0.035422 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 0.158967)
##
## Null deviance: 210.00 on 999 degrees of freedom
## Residual deviance: 155.79 on 980 degrees of freedom
## AIC: 1020.6
##
## Number of Fisher Scoring iterations: 2## remove savingsA62
model7 <- glm(formula = classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
duration + credithistoryA30 + credithistoryA31 + credithistoryA32 +
purposeA40 + purposeA41 + purposeA46 +
savingsA61 + employmentlengthA74 +
installmentrate + marriagesexA93 + otherdebtorsA101 + otherdebtorsA102 +
otherinstallmentplansA141 + housingA151 +
foreignworker, data = creditDataNumeric)
summary(model7)##
## Call:
## glm(formula = classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
## duration + credithistoryA30 + credithistoryA31 + credithistoryA32 +
## purposeA40 + purposeA41 + purposeA46 + savingsA61 + employmentlengthA74 +
## installmentrate + marriagesexA93 + otherdebtorsA101 + otherdebtorsA102 +
## otherinstallmentplansA141 + housingA151 + foreignworker,
## data = creditDataNumeric)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.91266 -0.29074 -0.09822 0.32644 1.06082
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -0.396204 0.080086 -4.947 8.85e-07 ***
## existingcheckingstatusA11 0.268345 0.032557 8.242 5.38e-16 ***
## existingcheckingstatusA12 0.200092 0.031665 6.319 3.99e-10 ***
## duration 0.008337 0.001165 7.159 1.59e-12 ***
## credithistoryA30 0.271370 0.068309 3.973 7.63e-05 ***
## credithistoryA31 0.254393 0.063100 4.032 5.97e-05 ***
## credithistoryA32 0.084015 0.027500 3.055 0.00231 **
## purposeA40 0.132254 0.031404 4.211 2.77e-05 ***
## purposeA41 -0.102875 0.043945 -2.341 0.01943 *
## purposeA46 0.164758 0.059155 2.785 0.00545 **
## savingsA61 0.108008 0.027128 3.981 7.36e-05 ***
## employmentlengthA74 -0.087738 0.033621 -2.610 0.00920 **
## installmentrate 0.026099 0.011631 2.244 0.02506 *
## marriagesexA93 -0.060190 0.026853 -2.241 0.02522 *
## otherdebtorsA101 0.192226 0.058663 3.277 0.00109 **
## otherdebtorsA102 0.277756 0.084201 3.299 0.00101 **
## otherinstallmentplansA141 0.074747 0.037794 1.978 0.04824 *
## housingA151 0.081336 0.034383 2.366 0.01819 *
## foreignworker -0.145977 0.069320 -2.106 0.03547 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 0.1594485)
##
## Null deviance: 210.00 on 999 degrees of freedom
## Residual deviance: 156.42 on 981 degrees of freedom
## AIC: 1022.7
##
## Number of Fisher Scoring iterations: 2## remove otherinstallmentplansA141
model8 <- glm(formula = classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
duration + credithistoryA30 + credithistoryA31 + credithistoryA32 +
purposeA40 + purposeA41 + purposeA46 +
savingsA61 + employmentlengthA74 +
installmentrate + marriagesexA93 + otherdebtorsA101 + otherdebtorsA102 +
housingA151 +
foreignworker, data = creditDataNumeric)
summary(model8)##
## Call:
## glm(formula = classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
## duration + credithistoryA30 + credithistoryA31 + credithistoryA32 +
## purposeA40 + purposeA41 + purposeA46 + savingsA61 + employmentlengthA74 +
## installmentrate + marriagesexA93 + otherdebtorsA101 + otherdebtorsA102 +
## housingA151 + foreignworker, data = creditDataNumeric)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.9528 -0.2895 -0.1014 0.3169 1.0522
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -0.380278 0.079798 -4.765 2.17e-06 ***
## existingcheckingstatusA11 0.268497 0.032605 8.235 5.69e-16 ***
## existingcheckingstatusA12 0.201927 0.031698 6.370 2.89e-10 ***
## duration 0.008384 0.001166 7.190 1.28e-12 ***
## credithistoryA30 0.280788 0.068244 4.114 4.21e-05 ***
## credithistoryA31 0.281612 0.061672 4.566 5.59e-06 ***
## credithistoryA32 0.083689 0.027541 3.039 0.00244 **
## purposeA40 0.133062 0.031448 4.231 2.54e-05 ***
## purposeA41 -0.103782 0.044007 -2.358 0.01855 *
## purposeA46 0.167140 0.059230 2.822 0.00487 **
## savingsA61 0.107732 0.027168 3.965 7.86e-05 ***
## employmentlengthA74 -0.088162 0.033670 -2.618 0.00897 **
## installmentrate 0.025496 0.011645 2.190 0.02879 *
## marriagesexA93 -0.058776 0.026883 -2.186 0.02902 *
## otherdebtorsA101 0.184556 0.058621 3.148 0.00169 **
## otherdebtorsA102 0.270895 0.084254 3.215 0.00135 **
## housingA151 0.079765 0.034424 2.317 0.02070 *
## foreignworker -0.146605 0.069422 -2.112 0.03495 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 0.1599213)
##
## Null deviance: 210.00 on 999 degrees of freedom
## Residual deviance: 157.04 on 982 degrees of freedom
## AIC: 1024.6
##
## Number of Fisher Scoring iterations: 2## remove foreignworker
model9 <- glm(formula = classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
duration + credithistoryA30 + credithistoryA31 + credithistoryA32 +
purposeA40 + purposeA41 + purposeA46 +
savingsA61 + employmentlengthA74 +
installmentrate + marriagesexA93 + otherdebtorsA101 + otherdebtorsA102 +
housingA151, data = creditDataNumeric)
summary(model9)##
## Call:
## glm(formula = classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
## duration + credithistoryA30 + credithistoryA31 + credithistoryA32 +
## purposeA40 + purposeA41 + purposeA46 + savingsA61 + employmentlengthA74 +
## installmentrate + marriagesexA93 + otherdebtorsA101 + otherdebtorsA102 +
## housingA151, data = creditDataNumeric)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.9463 -0.2903 -0.1038 0.3177 1.0637
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -0.407342 0.078901 -5.163 2.95e-07 ***
## existingcheckingstatusA11 0.266196 0.032644 8.154 1.06e-15 ***
## existingcheckingstatusA12 0.204848 0.031723 6.457 1.67e-10 ***
## duration 0.008688 0.001159 7.495 1.48e-13 ***
## credithistoryA30 0.275824 0.068324 4.037 5.83e-05 ***
## credithistoryA31 0.284961 0.061760 4.614 4.47e-06 ***
## credithistoryA32 0.084134 0.027588 3.050 0.00235 **
## purposeA40 0.124456 0.031237 3.984 7.27e-05 ***
## purposeA41 -0.104042 0.044084 -2.360 0.01847 *
## purposeA46 0.168823 0.059329 2.846 0.00453 **
## savingsA61 0.109610 0.027201 4.030 6.02e-05 ***
## employmentlengthA74 -0.089166 0.033726 -2.644 0.00833 **
## installmentrate 0.027397 0.011630 2.356 0.01869 *
## marriagesexA93 -0.061381 0.026902 -2.282 0.02272 *
## otherdebtorsA101 0.197665 0.058394 3.385 0.00074 ***
## otherdebtorsA102 0.274102 0.084389 3.248 0.00120 **
## housingA151 0.078315 0.034478 2.271 0.02333 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 0.1604841)
##
## Null deviance: 210.00 on 999 degrees of freedom
## Residual deviance: 157.76 on 983 degrees of freedom
## AIC: 1027.2
##
## Number of Fisher Scoring iterations: 2## remove housingA151
model10 <- glm(formula = classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
duration + credithistoryA30 + credithistoryA31 + credithistoryA32 +
purposeA40 + purposeA41 + purposeA46 +
savingsA61 + employmentlengthA74 +
installmentrate + marriagesexA93 + otherdebtorsA101 + otherdebtorsA102, data = creditDataNumeric)
summary(model10)##
## Call:
## glm(formula = classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
## duration + credithistoryA30 + credithistoryA31 + credithistoryA32 +
## purposeA40 + purposeA41 + purposeA46 + savingsA61 + employmentlengthA74 +
## installmentrate + marriagesexA93 + otherdebtorsA101 + otherdebtorsA102,
## data = creditDataNumeric)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.9570 -0.2941 -0.1049 0.3235 1.0546
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -0.381250 0.078225 -4.874 1.28e-06 ***
## existingcheckingstatusA11 0.273263 0.032564 8.392 < 2e-16 ***
## existingcheckingstatusA12 0.206362 0.031783 6.493 1.33e-10 ***
## duration 0.008534 0.001160 7.360 3.89e-13 ***
## credithistoryA30 0.285911 0.068323 4.185 3.11e-05 ***
## credithistoryA31 0.288335 0.061872 4.660 3.59e-06 ***
## credithistoryA32 0.087859 0.027598 3.184 0.001500 **
## purposeA40 0.124033 0.031303 3.962 7.96e-05 ***
## purposeA41 -0.097373 0.044079 -2.209 0.027403 *
## purposeA46 0.171101 0.059446 2.878 0.004085 **
## savingsA61 0.106760 0.027230 3.921 9.44e-05 ***
## employmentlengthA74 -0.086843 0.033782 -2.571 0.010295 *
## installmentrate 0.025920 0.011636 2.227 0.026140 *
## marriagesexA93 -0.074283 0.026351 -2.819 0.004914 **
## otherdebtorsA101 0.195497 0.058509 3.341 0.000865 ***
## otherdebtorsA102 0.280463 0.084520 3.318 0.000939 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 0.1611625)
##
## Null deviance: 210.00 on 999 degrees of freedom
## Residual deviance: 158.58 on 984 degrees of freedom
## AIC: 1030.4
##
## Number of Fisher Scoring iterations: 2## remove purposeA41
model11 <- glm(formula = classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
duration + credithistoryA30 + credithistoryA31 + credithistoryA32 +
purposeA40 + purposeA46 +
savingsA61 + employmentlengthA74 +
installmentrate + marriagesexA93 + otherdebtorsA101 + otherdebtorsA102, data = creditDataNumeric)
summary(model11)##
## Call:
## glm(formula = classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
## duration + credithistoryA30 + credithistoryA31 + credithistoryA32 +
## purposeA40 + purposeA46 + savingsA61 + employmentlengthA74 +
## installmentrate + marriagesexA93 + otherdebtorsA101 + otherdebtorsA102,
## data = creditDataNumeric)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.9577 -0.2943 -0.1020 0.3270 1.0739
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -0.396966 0.078054 -5.086 4.38e-07 ***
## existingcheckingstatusA11 0.273048 0.032628 8.368 < 2e-16 ***
## existingcheckingstatusA12 0.210577 0.031788 6.624 5.73e-11 ***
## duration 0.008167 0.001150 7.103 2.34e-12 ***
## credithistoryA30 0.294260 0.068353 4.305 1.84e-05 ***
## credithistoryA31 0.289283 0.061992 4.666 3.49e-06 ***
## credithistoryA32 0.089126 0.027646 3.224 0.001306 **
## purposeA40 0.137755 0.030741 4.481 8.29e-06 ***
## purposeA46 0.185465 0.059205 3.133 0.001784 **
## savingsA61 0.111127 0.027211 4.084 4.79e-05 ***
## employmentlengthA74 -0.085691 0.033844 -2.532 0.011499 *
## installmentrate 0.029352 0.011555 2.540 0.011231 *
## marriagesexA93 -0.079455 0.026298 -3.021 0.002582 **
## otherdebtorsA101 0.191918 0.058602 3.275 0.001094 **
## otherdebtorsA102 0.287047 0.084634 3.392 0.000722 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 0.1617973)
##
## Null deviance: 210.00 on 999 degrees of freedom
## Residual deviance: 159.37 on 985 degrees of freedom
## AIC: 1033.4
##
## Number of Fisher Scoring iterations: 2## remove employmentlengthA74
model12 <- glm(formula = classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
duration + credithistoryA30 + credithistoryA31 + credithistoryA32 +
purposeA40 + purposeA46 +
savingsA61 +
installmentrate + marriagesexA93 + otherdebtorsA101 + otherdebtorsA102, data = creditDataNumeric)
summary(model12)##
## Call:
## glm(formula = classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
## duration + credithistoryA30 + credithistoryA31 + credithistoryA32 +
## purposeA40 + purposeA46 + savingsA61 + installmentrate +
## marriagesexA93 + otherdebtorsA101 + otherdebtorsA102, data = creditDataNumeric)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.9352 -0.2957 -0.1051 0.3316 1.0910
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -0.412170 0.078036 -5.282 1.57e-07 ***
## existingcheckingstatusA11 0.273803 0.032716 8.369 < 2e-16 ***
## existingcheckingstatusA12 0.211449 0.031874 6.634 5.38e-11 ***
## duration 0.007919 0.001149 6.893 9.74e-12 ***
## credithistoryA30 0.305557 0.068394 4.468 8.82e-06 ***
## credithistoryA31 0.288162 0.062161 4.636 4.03e-06 ***
## credithistoryA32 0.090224 0.027718 3.255 0.001172 **
## purposeA40 0.136854 0.030823 4.440 1.00e-05 ***
## purposeA46 0.188729 0.059353 3.180 0.001520 **
## savingsA61 0.112216 0.027282 4.113 4.23e-05 ***
## installmentrate 0.029868 0.011585 2.578 0.010075 *
## marriagesexA93 -0.083124 0.026330 -3.157 0.001643 **
## otherdebtorsA101 0.195864 0.058742 3.334 0.000887 ***
## otherdebtorsA102 0.294405 0.084816 3.471 0.000541 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 0.1626852)
##
## Null deviance: 210.00 on 999 degrees of freedom
## Residual deviance: 160.41 on 986 degrees of freedom
## AIC: 1037.8
##
## Number of Fisher Scoring iterations: 2## remove installmentrate
model13 <- glm(formula = classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
duration + credithistoryA30 + credithistoryA31 + credithistoryA32 +
purposeA40 + purposeA46 +
savingsA61 +
marriagesexA93 + otherdebtorsA101 + otherdebtorsA102, data = creditDataNumeric)
summary(model13)##
## Call:
## glm(formula = classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
## duration + credithistoryA30 + credithistoryA31 + credithistoryA32 +
## purposeA40 + purposeA46 + savingsA61 + marriagesexA93 + otherdebtorsA101 +
## otherdebtorsA102, data = creditDataNumeric)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.9083 -0.2838 -0.1014 0.3361 1.1126
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -0.330578 0.071535 -4.621 4.32e-06 ***
## existingcheckingstatusA11 0.276911 0.032787 8.446 < 2e-16 ***
## existingcheckingstatusA12 0.209030 0.031951 6.542 9.72e-11 ***
## duration 0.008128 0.001149 7.072 2.89e-12 ***
## credithistoryA30 0.294359 0.068451 4.300 1.87e-05 ***
## credithistoryA31 0.290935 0.062329 4.668 3.47e-06 ***
## credithistoryA32 0.089435 0.027796 3.218 0.001335 **
## purposeA40 0.133545 0.030884 4.324 1.69e-05 ***
## purposeA46 0.193724 0.059491 3.256 0.001167 **
## savingsA61 0.111680 0.027359 4.082 4.83e-05 ***
## marriagesexA93 -0.075803 0.026252 -2.888 0.003967 **
## otherdebtorsA101 0.196291 0.058909 3.332 0.000894 ***
## otherdebtorsA102 0.291052 0.085048 3.422 0.000647 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 0.163616)
##
## Null deviance: 210.00 on 999 degrees of freedom
## Residual deviance: 161.49 on 987 degrees of freedom
## AIC: 1042.6
##
## Number of Fisher Scoring iterations: 2## credithistoryA32
model14 <- glm(formula = classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
duration + credithistoryA30 + credithistoryA31 +
purposeA40 + purposeA46 +
savingsA61 +
marriagesexA93 + otherdebtorsA101 + otherdebtorsA102, data = creditDataNumeric)
summary(model14)##
## Call:
## glm(formula = classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
## duration + credithistoryA30 + credithistoryA31 + purposeA40 +
## purposeA46 + savingsA61 + marriagesexA93 + otherdebtorsA101 +
## otherdebtorsA102, data = creditDataNumeric)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.9018 -0.2965 -0.1114 0.3467 1.1388
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -0.261331 0.068543 -3.813 0.000146 ***
## existingcheckingstatusA11 0.290420 0.032671 8.889 < 2e-16 ***
## existingcheckingstatusA12 0.216919 0.032007 6.777 2.10e-11 ***
## duration 0.007982 0.001154 6.918 8.23e-12 ***
## credithistoryA30 0.242052 0.066806 3.623 0.000306 ***
## credithistoryA31 0.235710 0.060202 3.915 9.65e-05 ***
## purposeA40 0.129431 0.031003 4.175 3.25e-05 ***
## purposeA46 0.185517 0.059717 3.107 0.001946 **
## savingsA61 0.107411 0.027456 3.912 9.78e-05 ***
## marriagesexA93 -0.086718 0.026154 -3.316 0.000948 ***
## otherdebtorsA101 0.185641 0.059094 3.141 0.001731 **
## otherdebtorsA102 0.284701 0.085427 3.333 0.000892 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 0.1651648)
##
## Null deviance: 210.00 on 999 degrees of freedom
## Residual deviance: 163.18 on 988 degrees of freedom
## AIC: 1051
##
## Number of Fisher Scoring iterations: 2## remove purposeA46
model15 <- glm(formula = classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
duration + credithistoryA30 + credithistoryA31 +
purposeA40 +
savingsA61 +
marriagesexA93 + otherdebtorsA101 + otherdebtorsA102, data = creditDataNumeric)
summary(model15)##
## Call:
## glm(formula = classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
## duration + credithistoryA30 + credithistoryA31 + purposeA40 +
## savingsA61 + marriagesexA93 + otherdebtorsA101 + otherdebtorsA102,
## data = creditDataNumeric)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.8940 -0.2983 -0.1034 0.3553 1.1494
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -0.259465 0.068839 -3.769 0.000174 ***
## existingcheckingstatusA11 0.291363 0.032812 8.880 < 2e-16 ***
## existingcheckingstatusA12 0.217625 0.032146 6.770 2.21e-11 ***
## duration 0.007953 0.001159 6.863 1.19e-11 ***
## credithistoryA30 0.230900 0.067001 3.446 0.000592 ***
## credithistoryA31 0.238298 0.060459 3.941 8.67e-05 ***
## purposeA40 0.116962 0.030877 3.788 0.000161 ***
## savingsA61 0.107957 0.027576 3.915 9.66e-05 ***
## marriagesexA93 -0.086411 0.026268 -3.290 0.001039 **
## otherdebtorsA101 0.196978 0.059239 3.325 0.000916 ***
## otherdebtorsA102 0.285231 0.085799 3.324 0.000919 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 0.1666096)
##
## Null deviance: 210.00 on 999 degrees of freedom
## Residual deviance: 164.78 on 989 degrees of freedom
## AIC: 1058.7
##
## Number of Fisher Scoring iterations: 2## marriagesexA93
model16 <- glm(formula = classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
duration + credithistoryA30 + credithistoryA31 +
purposeA40 +
savingsA61 +
otherdebtorsA101 + otherdebtorsA102, data = creditDataNumeric)
summary(model16)##
## Call:
## glm(formula = classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
## duration + credithistoryA30 + credithistoryA31 + purposeA40 +
## savingsA61 + otherdebtorsA101 + otherdebtorsA102, data = creditDataNumeric)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.91272 -0.30089 -0.09708 0.37664 1.11610
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -0.303005 0.067889 -4.463 9.00e-06 ***
## existingcheckingstatusA11 0.295843 0.032946 8.980 < 2e-16 ***
## existingcheckingstatusA12 0.226177 0.032199 7.024 4.00e-12 ***
## duration 0.007455 0.001155 6.457 1.67e-10 ***
## credithistoryA30 0.226845 0.067321 3.370 0.000782 ***
## credithistoryA31 0.236330 0.060755 3.890 0.000107 ***
## purposeA40 0.112360 0.030997 3.625 0.000304 ***
## savingsA61 0.110455 0.027702 3.987 7.17e-05 ***
## otherdebtorsA101 0.200170 0.059524 3.363 0.000801 ***
## otherdebtorsA102 0.283037 0.086221 3.283 0.001064 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 0.1682624)
##
## Null deviance: 210.00 on 999 degrees of freedom
## Residual deviance: 166.58 on 990 degrees of freedom
## AIC: 1067.6
##
## Number of Fisher Scoring iterations: 2## otherdebtorsA102
model17 <- glm(formula = classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
duration + credithistoryA30 + credithistoryA31 +
purposeA40 +
savingsA61 +
otherdebtorsA101, data = creditDataNumeric)
summary(model17)##
## Call:
## glm(formula = classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
## duration + credithistoryA30 + credithistoryA31 + purposeA40 +
## savingsA61 + otherdebtorsA101, data = creditDataNumeric)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.8212 -0.2903 -0.1094 0.3845 1.0105
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -0.175993 0.056059 -3.139 0.001743 **
## existingcheckingstatusA11 0.295065 0.033107 8.912 < 2e-16 ***
## existingcheckingstatusA12 0.217313 0.032244 6.740 2.69e-11 ***
## duration 0.007662 0.001159 6.614 6.11e-11 ***
## credithistoryA30 0.224475 0.067648 3.318 0.000939 ***
## credithistoryA31 0.232596 0.061044 3.810 0.000147 ***
## purposeA40 0.114032 0.031146 3.661 0.000264 ***
## savingsA61 0.106454 0.027811 3.828 0.000137 ***
## otherdebtorsA101 0.073584 0.045567 1.615 0.106659
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 0.1699222)
##
## Null deviance: 210.00 on 999 degrees of freedom
## Residual deviance: 168.39 on 991 degrees of freedom
## AIC: 1076.4
##
## Number of Fisher Scoring iterations: 2## otherdebtorsA101
model18 <- glm(formula = classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
duration + credithistoryA30 + credithistoryA31 +
purposeA40 +
savingsA61, data = creditDataNumeric)
summary(model18)##
## Call:
## glm(formula = classification ~ existingcheckingstatusA11 + existingcheckingstatusA12 +
## duration + credithistoryA30 + credithistoryA31 + purposeA40 +
## savingsA61, data = creditDataNumeric)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.8221 -0.2917 -0.1129 0.3892 1.0126
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -0.104863 0.034704 -3.022 0.002579 **
## existingcheckingstatusA11 0.290386 0.033007 8.798 < 2e-16 ***
## existingcheckingstatusA12 0.212121 0.032109 6.606 6.42e-11 ***
## duration 0.007685 0.001159 6.629 5.54e-11 ***
## credithistoryA30 0.231170 0.067576 3.421 0.000650 ***
## credithistoryA31 0.228067 0.061028 3.737 0.000197 ***
## purposeA40 0.114729 0.031168 3.681 0.000245 ***
## savingsA61 0.102479 0.027724 3.696 0.000231 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for gaussian family taken to be 0.1701976)
##
## Null deviance: 210.00 on 999 degrees of freedom
## Residual deviance: 168.84 on 992 degrees of freedom
## AIC: 1077
##
## Number of Fisher Scoring iterations: 2#### at this point all the variables are quite significant so chhosing this as final glm model
glm_final <- model18
## predict using glm_final
test_pred = predict(glm_final, type = "response",
newdata = test[,-60])
test$pred <- test_pred
summary(test$pred)## Min. 1st Qu. Median Mean 3rd Qu. Max.
## -0.05875 0.12556 0.29554 0.30805 0.47245 0.86502## use a cutoff of 0.5
test_pred_default <- factor(ifelse(test_pred >= 0.50, "Yes", "No"))
test_actual_default <- factor(ifelse(test$classification==1,"Yes","No"))
## create table
table(test_actual_default,test_pred_default)## test_pred_default
## test_actual_default No Yes
## No 209 19
## Yes 36 36## confusion matrix
test_conf <- confusionMatrix(test_pred_default, test_actual_default, positive = "Yes")
test_conf## Confusion Matrix and Statistics
##
## Reference
## Prediction No Yes
## No 209 36
## Yes 19 36
##
## Accuracy : 0.8167
## 95% CI : (0.7682, 0.8588)
## No Information Rate : 0.76
## P-Value [Acc > NIR] : 0.01119
##
## Kappa : 0.4533
## Mcnemar's Test P-Value : 0.03097
##
## Sensitivity : 0.5000
## Specificity : 0.9167
## Pos Pred Value : 0.6545
## Neg Pred Value : 0.8531
## Prevalence : 0.2400
## Detection Rate : 0.1200
## Detection Prevalence : 0.1833
## Balanced Accuracy : 0.7083
##
## 'Positive' Class : Yes
## ## calculate cost
cost_glm <- sum(test_conf$table * as.vector(penaltyMatrix))
cost_glm## [1] 131## rf model
creditDataNumeric$classification <- as.factor(creditDataNumeric$classification)
## test trian split
set.seed(100)
indices = sample(1:nrow(creditDataNumeric), 0.7*nrow(creditDataNumeric))
train = creditDataNumeric[indices,]
test = creditDataNumeric[-indices,]
## rf model 1
model_rf1 <- randomForest(classification~.,data = train)
prediction <- predict(model_rf1, newdata = test[, -60], type = "response")
## accuracy
table(test$classification, prediction)## prediction
## 0 1
## 0 203 25
## 1 41 31## rf model 2
model_rf2 <- randomForest(classification~., data = train, cp = 0.05, parms = list(loss = PenaltyMatrix))
prediction2 <- predict(model_rf2, newdata = test[,-60], type = "response")
table(test$classification, prediction2)## prediction2
## 0 1
## 0 205 23
## 1 42 30plot(model_rf2)
## choose mtry
mtry <- tuneRF(train[,-60],train$classification,ntreeTry = 500, stepFactor = 1.5, improve = 0.001, trace = TRUE, plot = TRUE, doBest = TRUE)## mtry = 7 OOB error = 26.14%
## Searching left ...
## mtry = 5 OOB error = 25.29%
## 0.03278689 0.001
## mtry = 4 OOB error = 25.14%
## 0.005649718 0.001
## mtry = 3 OOB error = 25.86%
## -0.02840909 0.001
## Searching right ...
## mtry = 10 OOB error = 25.14%
## 0 0.001
## mtry = 4 as best mtry
## rf model 3
model_rf3 <- randomForest(classification~., data = train, cp = 0.05, parms = list(loss = PenaltyMatrix), mtry = 4, importance = TRUE)
print(model_rf3)##
## Call:
## randomForest(formula = classification ~ ., data = train, cp = 0.05, parms = list(loss = PenaltyMatrix), mtry = 4, importance = TRUE)
## Type of random forest: classification
## Number of trees: 500
## No. of variables tried at each split: 4
##
## OOB estimate of error rate: 25%
## Confusion matrix:
## 0 1 class.error
## 0 436 36 0.07627119
## 1 139 89 0.60964912## variable importance
importance(model_rf3)## 0 1 MeanDecreaseAccuracy
## existingcheckingstatusA11 9.080692126 14.28121679 15.0570381
## existingcheckingstatusA12 2.786507459 2.86949133 4.4982225
## existingcheckingstatusA13 3.279749844 -0.40912045 2.4726229
## existingcheckingstatusA14 12.727417815 13.65288494 17.4511119
## duration 11.743540745 5.61811979 12.3582962
## credithistoryA30 2.320588369 1.97847880 2.7097952
## credithistoryA31 6.107361546 2.88471262 6.2700321
## credithistoryA32 5.291696758 -2.36610958 3.3462744
## credithistoryA33 2.935872777 -0.31721073 2.1463237
## credithistoryA34 10.076248484 6.64052421 11.9705260
## purposeA40 2.417147586 0.50046933 2.2281303
## purposeA41 1.126105091 2.35542457 2.3702461
## purposeA410 0.846351238 -0.24354691 0.5804699
## purposeA42 3.633228668 -0.47342757 2.7566791
## purposeA43 1.612270587 0.69538597 1.7600515
## purposeA44 0.626004652 -0.70173223 0.1241783
## purposeA45 -1.158756892 -0.32522131 -1.0908808
## purposeA46 1.353800693 -0.76924239 0.4961081
## purposeA48 -0.793349060 -1.73248097 -1.5090550
## purposeA49 -0.497617482 2.11061743 0.8326941
## creditamount 9.035690179 3.75542130 9.8236961
## savingsA61 3.182878305 5.84777479 5.7176155
## savingsA62 1.321994583 1.58003114 2.0914117
## savingsA63 -0.157089763 1.88082168 0.8831813
## savingsA64 -1.143004494 3.43678238 0.8468561
## savingsA65 3.269037731 0.60142924 3.0051882
## employmentlengthA71 2.218892059 -0.47907818 1.6148397
## employmentlengthA72 1.379971614 0.09235138 1.1622822
## employmentlengthA73 0.522762849 -1.35392717 -0.3504567
## employmentlengthA74 2.672328954 0.46368920 2.4638264
## employmentlengthA75 -1.857591142 1.29292775 -0.7928400
## installmentrate 3.497652001 1.07299439 3.6508288
## marriagesexA91 -0.834682470 -0.12460335 -0.7455186
## marriagesexA92 1.098102886 2.88773873 3.0058791
## marriagesexA93 0.365179153 2.43308036 1.9390932
## marriagesexA94 1.783976912 -1.74968927 0.7052991
## otherdebtorsA101 2.774125974 -0.31797703 2.1921775
## otherdebtorsA102 -0.271698962 0.98663312 0.5116362
## otherdebtorsA103 6.169815495 1.40678755 5.6353646
## presentresidencelength 0.003164577 -0.76765976 -0.4273986
## propertyA121 5.762761234 3.51932843 6.7791129
## propertyA122 1.890211207 -0.13243604 1.4388813
## propertyA123 2.567236668 -0.23103202 1.8626013
## propertyA124 4.646296412 0.38328981 4.2235963
## age 2.705513923 0.92650383 2.8417631
## otherinstallmentplansA141 4.845457550 2.27318861 5.2499267
## otherinstallmentplansA142 1.249168623 0.32254949 1.1190379
## otherinstallmentplansA143 6.173699553 1.38157284 6.1855076
## housingA151 1.058937584 -1.74763002 -0.3033677
## housingA152 3.683871264 1.73664957 4.0882695
## housingA153 2.195464875 -0.28451854 1.6423698
## existingcredits 3.159581353 0.62225915 3.1675168
## jobA171 0.290548823 -0.60817998 -0.1522321
## jobA172 2.827683659 -1.52236621 1.4163805
## jobA173 0.886712317 1.91059840 1.8438825
## jobA174 -0.062388557 3.99417946 2.6082871
## peopleliableno 2.088698422 -1.47768892 0.7919478
## telephone 1.894656466 1.49985492 2.4100477
## foreignworker 0.649112825 2.62577303 1.9177416
## MeanDecreaseGini
## existingcheckingstatusA11 10.9459486
## existingcheckingstatusA12 4.4564143
## existingcheckingstatusA13 2.1869305
## existingcheckingstatusA14 12.8075596
## duration 19.1635472
## credithistoryA30 1.9279234
## credithistoryA31 2.7079246
## credithistoryA32 3.9875898
## credithistoryA33 2.8133356
## credithistoryA34 7.1774770
## purposeA40 4.0764311
## purposeA41 1.9541701
## purposeA410 0.5443075
## purposeA42 3.5323458
## purposeA43 4.1791943
## purposeA44 0.6634778
## purposeA45 1.1328302
## purposeA46 2.8404772
## purposeA48 0.2893075
## purposeA49 2.5350891
## creditamount 24.1711808
## savingsA61 5.2903305
## savingsA62 2.3931888
## savingsA63 1.7741433
## savingsA64 1.5612315
## savingsA65 3.4190278
## employmentlengthA71 2.2879661
## employmentlengthA72 3.4030971
## employmentlengthA73 4.0809854
## employmentlengthA74 3.3107425
## employmentlengthA75 3.5445937
## installmentrate 8.1006096
## marriagesexA91 1.7748568
## marriagesexA92 4.6023113
## marriagesexA93 3.8921136
## marriagesexA94 2.0842178
## otherdebtorsA101 2.4224973
## otherdebtorsA102 2.1530840
## otherdebtorsA103 1.8899214
## presentresidencelength 8.2411203
## propertyA121 4.6249471
## propertyA122 3.2071219
## propertyA123 3.9306645
## propertyA124 3.7631834
## age 19.2919484
## otherinstallmentplansA141 3.8214752
## otherinstallmentplansA142 1.8435011
## otherinstallmentplansA143 4.6545436
## housingA151 3.1562046
## housingA152 3.9878777
## housingA153 2.0143343
## existingcredits 4.0107119
## jobA171 0.8451022
## jobA172 3.3113921
## jobA173 4.1923224
## jobA174 3.2002755
## peopleliableno 3.0897505
## telephone 4.5292018
## foreignworker 0.8547011varImpPlot(model_rf3)
## ROC & auc
predrf <- predict(model_rf3, newdata = test[,-60], type = "prob")
pred <- prediction(predrf[,2], test$classification)
# roc
prediction <- performance(pred, "tpr", "fpr")
plot(prediction,main="ROC Curve for Random Forest",col=2,lwd=2)
# auc
auc <- as.numeric(performance(pred, "auc")@y.values)
auc## [1] 0.7869761##
test_pred_default_rf = predict(model_rf3, type = "response",
newdata = test[,-60])
test_actual_default_rf <- test$classification
## confusion matrix
test_conf_rf <- confusionMatrix(test_pred_default_rf, test_actual_default_rf, positive = "1")
## calculate cost
cost_rf <- sum(test_conf_rf$table * as.vector(penaltyMatrix))
cost_rf## [1] 153#### glm one is better model with lower cost
final_model <- glm_final
creditDataNumeric$prediction <- predict(final_model, type = "response", newdata = creditDataNumeric[, -60])
creditDataNumeric$prediction <- factor(ifelse(creditDataNumeric$prediction >= 0.50, "Default", "Not Default"))
creditData$prediction <- creditDataNumeric$prediction