Assignment (Case: Auto Finance)
Sameer Mathur
Reading & Explaining the Data
Please download the data using the link
Please review the code below and execute it in your own RMD file.
Reading the Data
# reading the data
df <- read.csv("AutoFinanaceData.csv")
# attach the data frame
attach(df)
# Number of rows and columns
dim(df)## [1] 28906 21Printing the Column Names of the Data
## [1] "Agmt.No" "ContractStatus" "StartDate" "AGE"
## [5] "NOOFDEPE" "MTHINCTH" "SALDATFR" "TENORYR"
## [9] "DWNPMFR" "PROFBUS" "QUALHSC" "QUAL_PG"
## [13] "SEXCODE" "FULLPDC" "FRICODE" "WASHCODE"
## [17] "Region" "Branch" "DefaulterFlag" "DefaulterType"
## [21] "DATASET"List of Data Columns
DEFAULT
1. Defaulter Flag
1: Customer has delayed paying at least once
0: Otherwise
DEMOGRAPHIC VARIABLES
1. Gender
SEXCODE = 1 (Male)
SEXCODE = 0 (Female)
2. Age
3. Education
QUALHSC
QUAL_PG
4. Income
Monthly Income in Thousands (MTHINCTH)
Owns a Fridge (FRICODE)
Owns a Washing Machine (WASHCODE)
5. Profession
- PROFBUS = 1 (BUSINESS)
- PROFBUS = 0 (PROFESSIONAL)
6. No.of Dependents
- NOOFDEPE
7. Region
Structure of the Dataset
## 'data.frame': 28906 obs. of 21 variables:
## $ Agmt.No : chr "AP18100057" "AP18100140" "AP18100198" "AP18100217" ...
## $ ContractStatus: chr "Closed" "Closed" "Closed" "Closed" ...
## $ StartDate : chr "19-01-01" "10-05-01" "05-08-01" "03-09-01" ...
## $ AGE : int 26 28 32 31 36 33 41 47 43 27 ...
## $ NOOFDEPE : int 2 2 2 0 2 2 2 0 0 0 ...
## $ MTHINCTH : num 4.5 5.59 8.8 5 12 ...
## $ SALDATFR : num 1 1 1 1 1 1 1 1 0.97 1 ...
## $ TENORYR : num 1.5 2 1 1 1 2 1 2 1.5 2 ...
## $ DWNPMFR : num 0.27 0.25 0.51 0.66 0.17 0.18 0.37 0.42 0.27 0.47 ...
## $ PROFBUS : int 0 0 0 0 0 0 0 0 0 0 ...
## $ QUALHSC : int 0 0 0 0 0 0 1 0 0 0 ...
## $ QUAL_PG : int 0 0 0 0 0 0 0 0 0 0 ...
## $ SEXCODE : int 1 1 1 1 1 1 1 1 1 1 ...
## $ FULLPDC : int 1 1 1 1 1 0 0 1 1 1 ...
## $ FRICODE : int 0 1 1 1 1 0 0 0 0 0 ...
## $ WASHCODE : int 0 0 1 1 0 0 0 0 0 0 ...
## $ Region : chr "AP2" "AP2" "AP2" "AP2" ...
## $ Branch : chr "Vizag" "Vizag" "Vizag" "Vizag" ...
## $ DefaulterFlag : int 0 0 0 0 0 0 0 0 0 0 ...
## $ DefaulterType : int 0 0 0 0 0 0 0 0 0 0 ...
## $ DATASET : chr " " "BUILD" "BUILD" "BUILD" ...Convert catgorical variables to factor
## 'data.frame': 28906 obs. of 21 variables:
## $ Agmt.No : chr "AP18100057" "AP18100140" "AP18100198" "AP18100217" ...
## $ ContractStatus: chr "Closed" "Closed" "Closed" "Closed" ...
## $ StartDate : chr "19-01-01" "10-05-01" "05-08-01" "03-09-01" ...
## $ AGE : int 26 28 32 31 36 33 41 47 43 27 ...
## $ NOOFDEPE : int 2 2 2 0 2 2 2 0 0 0 ...
## $ MTHINCTH : num 4.5 5.59 8.8 5 12 ...
## $ SALDATFR : num 1 1 1 1 1 1 1 1 0.97 1 ...
## $ TENORYR : num 1.5 2 1 1 1 2 1 2 1.5 2 ...
## $ DWNPMFR : num 0.27 0.25 0.51 0.66 0.17 0.18 0.37 0.42 0.27 0.47 ...
## $ PROFBUS : Factor w/ 2 levels "0","1": 1 1 1 1 1 1 1 1 1 1 ...
## $ QUALHSC : Factor w/ 2 levels "0","1": 1 1 1 1 1 1 2 1 1 1 ...
## $ QUAL_PG : Factor w/ 2 levels "0","1": 1 1 1 1 1 1 1 1 1 1 ...
## $ SEXCODE : Factor w/ 2 levels "0","1": 2 2 2 2 2 2 2 2 2 2 ...
## $ FULLPDC : Factor w/ 2 levels "0","1": 2 2 2 2 2 1 1 2 2 2 ...
## $ FRICODE : Factor w/ 2 levels "0","1": 1 2 2 2 2 1 1 1 1 1 ...
## $ WASHCODE : Factor w/ 2 levels "0","1": 1 1 2 2 1 1 1 1 1 1 ...
## $ Region : Factor w/ 8 levels "AP1","AP2","Chennai",..: 2 2 2 2 2 2 2 2 2 2 ...
## $ Branch : Factor w/ 14 levels "Bangalore","Chennai",..: 14 14 14 14 14 14 14 14 14 14 ...
## $ DefaulterFlag : Factor w/ 2 levels "0","1": 1 1 1 1 1 1 1 1 1 1 ...
## $ DefaulterType : Factor w/ 3 levels "0","1","2": 1 1 1 1 1 1 1 1 1 1 ...
## $ DATASET : chr " " "BUILD" "BUILD" "BUILD" ...Descriptive Statistics
## n mean sd median min max
## Agmt.No* 28906 14453.50 8344.59 14453.50 1.00 28906.00
## ContractStatus* 28906 1.33 0.77 1.00 1.00 4.00
## StartDate* 28906 827.85 552.47 812.00 1.00 1814.00
## AGE 28906 36.44 9.82 35.00 18.00 70.00
## NOOFDEPE 28906 2.85 1.61 3.00 0.00 10.00
## MTHINCTH 28906 8.94 4.81 8.00 0.10 39.50
## SALDATFR 28906 0.44 0.46 0.17 0.03 1.03
## TENORYR 28906 1.28 0.52 1.00 0.17 4.00
## DWNPMFR 28906 0.38 0.16 0.38 0.02 0.88
## PROFBUS* 28906 1.15 0.36 1.00 1.00 2.00
## QUALHSC* 28906 1.23 0.42 1.00 1.00 2.00
## QUAL_PG* 28906 1.04 0.20 1.00 1.00 2.00
## SEXCODE* 28906 1.92 0.27 2.00 1.00 2.00
## FULLPDC* 28906 1.39 0.49 1.00 1.00 2.00
## FRICODE* 28906 1.42 0.49 1.00 1.00 2.00
## WASHCODE* 28906 1.19 0.39 1.00 1.00 2.00
## Region* 28906 5.33 1.51 6.00 1.00 8.00
## Branch* 28906 5.93 3.47 6.00 1.00 14.00
## DefaulterFlag* 28906 1.71 0.45 2.00 1.00 2.00
## DefaulterType* 28906 1.85 0.63 2.00 1.00 3.00
## DATASET* 28906 2.52 0.50 3.00 1.00 3.00Assignment: Part 2 (Logistic Regression)
Creating Train and Test dataset
Reserve 80% for training and 20% of test
## Warning: package 'caTools' was built under R version 4.0.4Que 1- Write R Code for Building The Logistics Regression Model using glm(), the output is as follows.
LogitModel <- glm(DefaulterFlag ~ AGE
+ NOOFDEPE
+ MTHINCTH
+ NOOFDEPE
+ SALDATFR
+ TENORYR
+ DWNPMFR
+ PROFBUS
+ QUALHSC
+ QUAL_PG
+ SEXCODE
+ FULLPDC
+ FRICODE
+ WASHCODE
+ Region,data = trainingSet, family = binomial())
summary(LogitModel)##
## Call:
## glm(formula = DefaulterFlag ~ AGE + NOOFDEPE + MTHINCTH + NOOFDEPE +
## SALDATFR + TENORYR + DWNPMFR + PROFBUS + QUALHSC + QUAL_PG +
## SEXCODE + FULLPDC + FRICODE + WASHCODE + Region, family = binomial(),
## data = trainingSet)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -2.7470 -1.0215 0.5716 0.7801 2.0874
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 2.2291732 0.1890284 11.793 < 2e-16 ***
## AGE -0.0144968 0.0016680 -8.691 < 2e-16 ***
## NOOFDEPE 0.0566860 0.0107923 5.252 1.50e-07 ***
## MTHINCTH -0.0004025 0.0035613 -0.113 0.910023
## SALDATFR -0.3833870 0.0420223 -9.123 < 2e-16 ***
## TENORYR 0.7727065 0.0456475 16.928 < 2e-16 ***
## DWNPMFR -1.3074501 0.1274734 -10.257 < 2e-16 ***
## PROFBUS1 0.1966576 0.0487903 4.031 5.56e-05 ***
## QUALHSC1 0.1853120 0.0401652 4.614 3.95e-06 ***
## QUAL_PG1 -0.2990904 0.0787907 -3.796 0.000147 ***
## SEXCODE1 0.2339445 0.0600322 3.897 9.74e-05 ***
## FULLPDC1 -1.2365885 0.0368674 -33.541 < 2e-16 ***
## FRICODE1 -0.1761473 0.0377247 -4.669 3.02e-06 ***
## WASHCODE1 -0.2644245 0.0476814 -5.546 2.93e-08 ***
## RegionAP2 -0.5788864 0.1796029 -3.223 0.001268 **
## RegionChennai -1.4136987 0.1408192 -10.039 < 2e-16 ***
## RegionKA1 -0.6529787 0.1411987 -4.625 3.75e-06 ***
## RegionKE2 -0.5753874 0.1450753 -3.966 7.30e-05 ***
## RegionTN1 -0.8084619 0.1362745 -5.933 2.98e-09 ***
## RegionTN2 -0.6142186 0.1458691 -4.211 2.55e-05 ***
## RegionVellore -0.6570233 0.1595604 -4.118 3.83e-05 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 26040 on 21678 degrees of freedom
## Residual deviance: 23025 on 21658 degrees of freedom
## AIC: 23067
##
## Number of Fisher Scoring iterations: 4Que 2- Write R Code for Building The Logistics Regression Model using caret Package, the output is as follows.
Take, set.seed(766)
library(caret)
# control parameters
objControl <- trainControl(method = "boot",
number = 2,
returnResamp = 'none',
summaryFunction = twoClassSummary,
classProbs = TRUE,
savePredictions = TRUE)
set.seed(766)
trainingSet$DefaulterFlag <- ifelse(trainingSet$DefaulterFlag == "1","Yes","No")
# model building using caret package
caretLogitModel <- train(trainingSet[,c(4:17)],
trainingSet$DefaulterFlag,
method = 'glm',
trControl = objControl,
metric = "ROC")
summary(caretLogitModel)##
## Call:
## NULL
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -2.7470 -1.0215 0.5716 0.7801 2.0874
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 2.2291732 0.1890284 11.793 < 2e-16 ***
## AGE -0.0144968 0.0016680 -8.691 < 2e-16 ***
## NOOFDEPE 0.0566860 0.0107923 5.252 1.50e-07 ***
## MTHINCTH -0.0004025 0.0035613 -0.113 0.910023
## SALDATFR -0.3833870 0.0420223 -9.123 < 2e-16 ***
## TENORYR 0.7727065 0.0456475 16.928 < 2e-16 ***
## DWNPMFR -1.3074501 0.1274734 -10.257 < 2e-16 ***
## PROFBUS1 0.1966576 0.0487903 4.031 5.56e-05 ***
## QUALHSC1 0.1853120 0.0401652 4.614 3.95e-06 ***
## QUAL_PG1 -0.2990904 0.0787907 -3.796 0.000147 ***
## SEXCODE1 0.2339445 0.0600322 3.897 9.74e-05 ***
## FULLPDC1 -1.2365885 0.0368674 -33.541 < 2e-16 ***
## FRICODE1 -0.1761473 0.0377247 -4.669 3.02e-06 ***
## WASHCODE1 -0.2644245 0.0476814 -5.546 2.93e-08 ***
## RegionAP2 -0.5788864 0.1796029 -3.223 0.001268 **
## RegionChennai -1.4136987 0.1408192 -10.039 < 2e-16 ***
## RegionKA1 -0.6529787 0.1411987 -4.625 3.75e-06 ***
## RegionKE2 -0.5753874 0.1450753 -3.966 7.30e-05 ***
## RegionTN1 -0.8084619 0.1362745 -5.933 2.98e-09 ***
## RegionTN2 -0.6142186 0.1458691 -4.211 2.55e-05 ***
## RegionVellore -0.6570233 0.1595604 -4.118 3.83e-05 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 26040 on 21678 degrees of freedom
## Residual deviance: 23025 on 21658 degrees of freedom
## AIC: 23067
##
## Number of Fisher Scoring iterations: 4Que 3- List the variable(s) which are not statistically significant in the Logistics Regression Model at 5% level of significance.
Que 4- Explain the impact of AGE on the Probability of Default in one or two sentences.
Que 5- Explain the impact of Gender (SEXCODE) on the Probability of Default in one or two sentences.
Que 6- Consider a consumer having the following characteristics:
Age = meanAge,
Male,
Education = UG,
MTHINCTH = mean(MTHINCTH)
NoOfDepe = mean(NOOFDEPE),
Owns a Fridge,
Owns a Washing Machine
Working Professional,
SALDATFR = mean(SALDATFR),
Lives in TN1
Tenure = mean(TENORYR)
Down Payment = mean(DWNPMFR) %
Did not submit FULLPDC
Please predict the probability of this consumer defaulting, based on the logisitic regression output
# creating single value data frame
Xdata <- data.frame(
AGE = 36,
NOOFDEPE = 3,
SALDATFR = 0,
TENORYR = 1.5,
DWNPMFR = 0.4,
PROFBUS = "0",
QUALHSC = "0",
QUAL_PG = "0",
SEXCODE = "0",
FULLPDC = "0",
FRICODE = "1",
WASHCODE = "1",
Region = "TN1",
MTHINCTH = mean(MTHINCTH))
pred1 <- predict(LogitModel, Xdata, type = "response")
pred1## 1
## 0.7791633Que 7- Write R code to predict the probabilities using test dataset.
Que 8- Write R Code to make the following Confusion Matrix assuming threshold Prob = 50%.
library(caret)
# confusion matrix using caret package
yPred <- ifelse(predProbTest > 0.5, "Yes", "No")
predY <- as.factor(yPred)
levels(testSet$DefaulterFlag) <- c("No", "Yes")
table(Prediction = predY, Actual = testSet$DefaulterFlag)## Actual
## Prediction No Yes
## No 635 417
## Yes 1448 4727Que 9- Calculate values of 3 Machine Learning Metrices (Accuarcy,Sensitivity,Specificity), and write R code to generate them.
## Confusion Matrix and Statistics
##
## Reference
## Prediction No Yes
## No 635 417
## Yes 1448 4727
##
## Accuracy : 0.7419
## 95% CI : (0.7317, 0.752)
## No Information Rate : 0.7118
## P-Value [Acc > NIR] : 5.682e-09
##
## Kappa : 0.2624
##
## Mcnemar's Test P-Value : < 2.2e-16
##
## Sensitivity : 0.9189
## Specificity : 0.3048
## Pos Pred Value : 0.7655
## Neg Pred Value : 0.6036
## Prevalence : 0.7118
## Detection Rate : 0.6541
## Detection Prevalence : 0.8544
## Balanced Accuracy : 0.6119
##
## 'Positive' Class : Yes
## Que 10- Write R code to draw this ROC Plot based on test dataset using the Logistic regression model.
## Warning: package 'ROCR' was built under R version 4.0.4# predicted probabilities
predProbTest <- predict(LogitModel, testSet, type = "response")
lgPredObj <- prediction(predProbTest,testSet$DefaulterFlag)
lgPerfObj <- performance(lgPredObj, "tpr","fpr")
plot(lgPerfObj,main = "ROC Curve",col = 2,lwd = 2)
abline(a = 0,b = 1,lwd = 2,lty = 3,col = "black")
Que 10- Write R code to to calculate area under curve (AUC), and write its implications in one or two sentences.
# auc for logit
aucLogit <- performance(lgPredObj, measure = "auc")
aucLogit <- aucLogit@y.values[[1]]
aucLogit## [1] 0.7263962