Data 621 Group 2 Final Project R Code
Mark Gonsalves, Joshua Hummell, Jiho Kim, Claire Meyer, Chinedu Onyeka, Rathish Parayil Sasidharan
4/17/2022
Load in data
bankruptcy <- read.csv("https://raw.githubusercontent.com/cmm6/data608/main/bankruptcy.csv", header=TRUE, sep = ",")
#bankruptcy <- read.csv("bankruptcy.csv", header=TRUE, sep = ",")Data Exploration
Then we’ll take a look at the dataset.
head(bankruptcy)## Bankrupt. ROA.C..before.interest.and.depreciation.before.interest
## 1 1 0.3705943
## 2 1 0.4642909
## 3 1 0.4260713
## 4 1 0.3998440
## 5 1 0.4650222
## 6 1 0.3886803
## ROA.A..before.interest.and...after.tax
## 1 0.4243894
## 2 0.5382141
## 3 0.4990188
## 4 0.4512647
## 5 0.5384322
## 6 0.4151766
## ROA.B..before.interest.and.depreciation.after.tax Operating.Gross.Margin
## 1 0.4057498 0.6014572
## 2 0.5167300 0.6102351
## 3 0.4722951 0.6014500
## 4 0.4577333 0.5835411
## 5 0.5222978 0.5987835
## 6 0.4191338 0.5901714
## Realized.Sales.Gross.Margin Operating.Profit.Rate Pre.tax.net.Interest.Rate
## 1 0.6014572 0.9989692 0.7968871
## 2 0.6102351 0.9989460 0.7973802
## 3 0.6013635 0.9988574 0.7964034
## 4 0.5835411 0.9986997 0.7969670
## 5 0.5987835 0.9989731 0.7973661
## 6 0.5902507 0.9987581 0.7969032
## After.tax.net.Interest.Rate Non.industry.income.and.expenditure.revenue
## 1 0.8088094 0.3026464
## 2 0.8093007 0.3035564
## 3 0.8083875 0.3020352
## 4 0.8089656 0.3033495
## 5 0.8093037 0.3034750
## 6 0.8087706 0.3031158
## Continuous.interest.rate..after.tax. Operating.Expense.Rate
## 1 0.7809849 1.256969e-04
## 2 0.7815060 2.897851e-04
## 3 0.7802839 2.361297e-04
## 4 0.7812410 1.078888e-04
## 5 0.7815500 7.890000e+09
## 6 0.7810691 1.571500e-04
## Research.and.development.expense.rate Cash.flow.rate
## 1 0 0.4581431
## 2 0 0.4618673
## 3 25500000 0.4585206
## 4 0 0.4657054
## 5 0 0.4627463
## 6 0 0.4658615
## Interest.bearing.debt.interest.rate Tax.rate..A. Net.Value.Per.Share..B.
## 1 0.0007250725 0 0.1479499
## 2 0.0006470647 0 0.1822511
## 3 0.0007900790 0 0.1779107
## 4 0.0004490449 0 0.1541865
## 5 0.0006860686 0 0.1675024
## 6 0.0007160716 0 0.1555771
## Net.Value.Per.Share..A. Net.Value.Per.Share..C.
## 1 0.1479499 0.1479499
## 2 0.1822511 0.1822511
## 3 0.1779107 0.1937129
## 4 0.1541865 0.1541865
## 5 0.1675024 0.1675024
## 6 0.1555771 0.1555771
## Persistent.EPS.in.the.Last.Four.Seasons Cash.Flow.Per.Share
## 1 0.1691406 0.3116644
## 2 0.2089439 0.3181368
## 3 0.1805805 0.3071019
## 4 0.1937222 0.3216736
## 5 0.2125366 0.3191625
## 6 0.1744351 0.3253873
## Revenue.Per.Share..Yuan.Â.. Operating.Profit.Per.Share..Yuan.Â..
## 1 0.017559780 0.09592053
## 2 0.021144335 0.09372201
## 3 0.005944008 0.09233776
## 4 0.014368468 0.07776240
## 5 0.029689792 0.09689765
## 6 0.018104270 0.07808810
## Per.Share.Net.profit.before.tax..Yuan.Â..
## 1 0.1387362
## 2 0.1699179
## 3 0.1428033
## 4 0.1486028
## 5 0.1684115
## 6 0.1388115
## Realized.Sales.Gross.Profit.Growth.Rate Operating.Profit.Growth.Rate
## 1 0.02210228 0.8481950
## 2 0.02208017 0.8480879
## 3 0.02276010 0.8480940
## 4 0.02204607 0.8480055
## 5 0.02209591 0.8482582
## 6 0.02156494 0.8479828
## After.tax.Net.Profit.Growth.Rate Regular.Net.Profit.Growth.Rate
## 1 0.6889795 0.6889795
## 2 0.6896929 0.6897017
## 3 0.6894627 0.6894697
## 4 0.6891095 0.6891095
## 5 0.6896969 0.6896969
## 6 0.6891051 0.6891775
## Continuous.Net.Profit.Growth.Rate Total.Asset.Growth.Rate
## 1 0.2175354 4.98e+09
## 2 0.2176196 6.11e+09
## 3 0.2176013 7.28e+09
## 4 0.2175682 4.88e+09
## 5 0.2176256 5.51e+09
## 6 0.2175664 6.08e+08
## Net.Value.Growth.Rate Total.Asset.Return.Growth.Rate.Ratio
## 1 0.0003269773 0.2631000
## 2 0.0004430401 0.2645158
## 3 0.0003964253 0.2641840
## 4 0.0003824259 0.2633712
## 5 0.0004389476 0.2652182
## 6 0.0003517819 0.2632500
## Cash.Reinvestment.. Current.Ratio Quick.Ratio Interest.Expense.Ratio
## 1 0.3637253 0.002258963 0.0012077551 0.6299513
## 2 0.3767091 0.006016206 0.0040393668 0.6351725
## 3 0.3689132 0.011542554 0.0053475602 0.6296314
## 4 0.3840766 0.004194059 0.0028964911 0.6302284
## 5 0.3796897 0.006022446 0.0037274466 0.6360550
## 6 0.3880258 0.002740085 0.0008546614 0.6301838
## Total.debt.Total.net.worth Debt.ratio.. Net.worth.Assets
## 1 0.021265924 0.2075763 0.7924237
## 2 0.012502394 0.1711763 0.8288237
## 3 0.021247686 0.2075158 0.7924842
## 4 0.009572402 0.1514648 0.8485352
## 5 0.005149600 0.1065091 0.8934909
## 6 0.014213152 0.1804275 0.8195725
## Long.term.fund.suitability.ratio..A. Borrowing.dependency
## 1 0.005024455 0.3902844
## 2 0.005058882 0.3767600
## 3 0.005099899 0.3790929
## 4 0.005046924 0.3797427
## 5 0.005303319 0.3750254
## 6 0.004913193 0.3814482
## Contingent.liabilities.Net.worth Operating.profit.Paid.in.capital
## 1 0.006478502 0.09588483
## 2 0.005835039 0.09374338
## 3 0.006561982 0.09231847
## 4 0.005365848 0.07772729
## 5 0.006623525 0.09692706
## 6 0.005749123 0.07810185
## Net.profit.before.tax.Paid.in.capital
## 1 0.1377573
## 2 0.1689616
## 3 0.1480356
## 4 0.1475605
## 5 0.1674610
## 6 0.1378252
## Inventory.and.accounts.receivable.Net.value Total.Asset.Turnover
## 1 0.3980357 0.08695652
## 2 0.3977249 0.06446777
## 3 0.4065805 0.01499250
## 4 0.3979245 0.08995502
## 5 0.4000788 0.17541229
## 6 0.4004191 0.09595202
## Accounts.Receivable.Turnover Average.Collection.Days
## 1 0.001813884 0.003487364
## 2 0.001286356 0.004916808
## 3 0.001495338 0.004226849
## 4 0.001966056 0.003214967
## 5 0.001448673 0.004366891
## 6 0.001527802 0.004137189
## Inventory.Turnover.Rate..times. Fixed.Assets.Turnover.Frequency
## 1 1.820926e-04 1.165007e-04
## 2 9.360000e+09 7.190000e+08
## 3 6.500000e+07 2.650000e+09
## 4 7.130000e+09 9.150000e+09
## 5 1.633674e-04 2.935211e-04
## 6 6.500000e+08 9.300000e+09
## Net.Worth.Turnover.Rate..times. Revenue.per.person
## 1 0.03290323 0.034164182
## 2 0.02548387 0.006888651
## 3 0.01338710 0.028996960
## 4 0.02806452 0.015463478
## 5 0.04016129 0.058111423
## 6 0.02967742 0.021300471
## Operating.profit.per.person Allocation.rate.per.person
## 1 0.3929129 0.03713530
## 2 0.3915900 0.01233497
## 3 0.3819678 0.14101631
## 4 0.3784966 0.02131999
## 5 0.3943715 0.02398821
## 6 0.3775695 0.03282877
## Working.Capital.to.Total.Assets Quick.Assets.Total.Assets
## 1 0.6727753 0.16667296
## 2 0.7511109 0.12723600
## 3 0.8295019 0.34020088
## 4 0.7257542 0.16157453
## 5 0.7518225 0.26032988
## 6 0.6867286 0.08026371
## Current.Assets.Total.Assets Cash.Total.Assets Quick.Assets.Current.Liability
## 1 0.1906430 0.0040944060 0.001996771
## 2 0.1824191 0.0149477270 0.004136030
## 3 0.6028057 0.0009909445 0.006302481
## 4 0.2258149 0.0188506248 0.002961238
## 5 0.3583802 0.0141609738 0.004274771
## 6 0.2145360 0.0026452256 0.000988425
## Cash.Current.Liability Current.Liability.to.Assets
## 1 1.473360e-04 0.14730845
## 2 1.383910e-03 0.05696283
## 3 5.340000e+09 0.09816206
## 4 1.010646e-03 0.09871463
## 5 6.804636e-04 0.11019485
## 6 1.008563e-04 0.13900211
## Operating.Funds.to.Liability Inventory.Working.Capital
## 1 0.3340152 0.2769202
## 2 0.3411060 0.2896416
## 3 0.3367315 0.2774555
## 4 0.3487164 0.2765803
## 5 0.3446388 0.2879127
## 6 0.3505631 0.2766785
## Inventory.Current.Liability Current.Liabilities.Liability
## 1 0.001035990 0.6762692
## 2 0.005209682 0.3085886
## 3 0.013878786 0.4460275
## 4 0.003540148 0.6158484
## 5 0.004868570 0.9750066
## 6 0.004879131 0.7333519
## Working.Capital.Equity Current.Liabilities.Equity
## 1 0.7212746 0.3390770
## 2 0.7319753 0.3297401
## 3 0.7427286 0.3347769
## 4 0.7298249 0.3315090
## 5 0.7319996 0.3307263
## 6 0.7252016 0.3355344
## Long.term.Liability.to.Current.Assets Retained.Earnings.to.Total.Assets
## 1 0.025592368 0.9032248
## 2 0.023946819 0.9310652
## 3 0.003715116 0.9099034
## 4 0.022165200 0.9069022
## 5 0.000000000 0.9138502
## 6 0.003772505 0.9030413
## Total.income.Total.expense Total.expense.Assets Current.Asset.Turnover.Rate
## 1 0.002021613 0.06485571 7.010000e+08
## 2 0.002225608 0.02551586 1.065198e-04
## 3 0.002060071 0.02138743 1.791094e-03
## 4 0.001831359 0.02416107 8.140000e+09
## 5 0.002223930 0.02638525 6.680000e+09
## 6 0.001865609 0.04009362 8.010000e+09
## Quick.Asset.Turnover.Rate Working.capitcal.Turnover.Rate Cash.Turnover.Rate
## 1 6.550000e+09 0.5938305 4.58e+08
## 2 7.700000e+09 0.5939155 2.49e+09
## 3 1.022676e-03 0.5945019 7.61e+08
## 4 6.050000e+09 0.5938888 2.03e+09
## 5 5.050000e+09 0.5939153 8.24e+08
## 6 2.810000e+09 0.5938458 2.95e+08
## Cash.Flow.to.Sales Fixed.Assets.to.Assets Current.Liability.to.Liability
## 1 0.6715677 0.4242058 0.6762692
## 2 0.6715699 0.4688281 0.3085886
## 3 0.6715713 0.2761792 0.4460275
## 4 0.6715192 0.5591440 0.6158484
## 5 0.6715631 0.3095549 0.9750066
## 6 0.6715676 0.6031935 0.7333519
## Current.Liability.to.Equity Equity.to.Long.term.Liability
## 1 0.3390770 0.1265495
## 2 0.3297401 0.1209161
## 3 0.3347769 0.1179223
## 4 0.3315090 0.1207605
## 5 0.3307263 0.1109332
## 6 0.3355344 0.1129172
## Cash.Flow.to.Total.Assets Cash.Flow.to.Liability CFO.to.Assets
## 1 0.6375554 0.4586091 0.5203819
## 2 0.6411000 0.4590011 0.5671013
## 3 0.6427646 0.4592540 0.5384905
## 4 0.5790393 0.4485179 0.6041051
## 5 0.6223741 0.4544109 0.5784689
## 6 0.6374698 0.4584993 0.6221901
## Cash.Flow.to.Equity Current.Liability.to.Current.Assets Liability.Assets.Flag
## 1 0.3129049 0.11825048 0
## 2 0.3141631 0.04777528 0
## 3 0.3145154 0.02534649 0
## 4 0.3023823 0.06724962 0
## 5 0.3115672 0.04772537 0
## 6 0.3132685 0.09952193 0
## Net.Income.to.Total.Assets Total.assets.to.GNP.price No.credit.Interval
## 1 0.7168453 0.009219440 0.6228790
## 2 0.7952971 0.008323302 0.6236517
## 3 0.7746697 0.040002853 0.6238410
## 4 0.7395545 0.003252475 0.6229287
## 5 0.7950159 0.003877563 0.6235207
## 6 0.7104205 0.005277875 0.6226046
## Gross.Profit.to.Sales Net.Income.to.Stockholder.s.Equity Liability.to.Equity
## 1 0.6014533 0.8278902 0.2902019
## 2 0.6102365 0.8399693 0.2838460
## 3 0.6014493 0.8367743 0.2901885
## 4 0.5835376 0.8346971 0.2817212
## 5 0.5987815 0.8399727 0.2785138
## 6 0.5901723 0.8299390 0.2850871
## Degree.of.Financial.Leverage..DFL.
## 1 0.02660063
## 2 0.26457682
## 3 0.02655472
## 4 0.02669663
## 5 0.02475185
## 6 0.02667537
## Interest.Coverage.Ratio..Interest.expense.to.EBIT. Net.Income.Flag
## 1 0.5640501 1
## 2 0.5701749 1
## 3 0.5637061 1
## 4 0.5646634 1
## 5 0.5756166 1
## 6 0.5645383 1
## Equity.to.Liability
## 1 0.01646874
## 2 0.02079431
## 3 0.01647411
## 4 0.02398233
## 5 0.03549020
## 6 0.01953448#summary(bankruptcy)
#is.na(bankruptcy)
#str(bankruptcy)First Impression :
The dataset has a combination of 6819 observations per each of 96 features. The data has 93 numerical features and 2 categorical features (Net.Income.Flag column only has one unique value). *The data looks clean without any missing data or null values.
Visualization
So we move on to plot some graphs to understand the data better to check for the distribution and potential outliers.
colnames(bankruptcy)[1] <- 'Bankrupt' #readbility purposeCorrelation
cor.matrix<-as.matrix(cor(bankruptcy))
correlation<-sort(cor.matrix[2:95],decreasing = T)
data1<-bankruptcy[,-95]
correlationMatrix <-cor(data1[,1:94])
hist1 = ggplot(bankruptcy, aes(Bankrupt,)) + geom_bar(fill=c("Green","Red")) + theme_bw()+ggtitle("Survive(0) VS Bankruptcy(1)")+theme(plot.title = element_text(hjust = 0.5))+xlab("Survive(0) Bankruptcy(1)")
corr1 = GGally::ggcorr(bankruptcy[,1:20],size=2,hjust=1)+geom_point()+scale_alpha_manual()+guides()## Registered S3 method overwritten by 'GGally':
## method from
## +.gg ggplot2corr2 = corrplot(corr=correlationMatrix, tl.cex=0.1)
The plot shows the correlation between each features.
Outliers
bankruptcy$Bankrupt <- factor(bankruptcy$Bankrupt , levels = c(0,1) ,labels = c('NO' , 'YES'))
bankruptcy %>% dplyr::select(where(is.numeric)) %>% multi.hist()
table(bankruptcy$Bankrupt)##
## NO YES
## 6599 220prop.table(table(bankruptcy$Bankrupt))##
## NO YES
## 0.9677372 0.0322628barplot(table(bankruptcy$Bankrupt),col = c("#eb8060", "#b9e38d"),
main="Frequency of Bankrutcy",
xlab="Bankruptcy Tag",
ylab="Number of Companies")
Out of the 6819 comapnies, 6599(97%) did not go bankrupt and 220(3%) went bankrupt. We see it’s highly class imbalanced - will want to play with downsampling/upsampling in some of our model development or we might have to overfit the data.
Data Preparation
From out exploration, Net.Income.Flag act as a noise so it will be dropped.
bankruptcy.clean <- bankruptcy[,-95]Normalize our data
normalize <- function(x){
return((x-min(x))/(max(x)-min(x)))
}
bankruptcy.clean <- as.data.frame(lapply(bankruptcy.clean[,2:94],normalize))
bankruptcy.clean <- as.data.frame(cbind(bankruptcy$Bankrupt, bankruptcy.clean))
colnames(bankruptcy.clean)[1] <- 'Bankrupt'Check to make sure there is no missing data
is_missing <- function(x){
missing_strs <- c('', 'null', 'na', 'nan', 'inf', '-inf', '-9', 'unknown', 'missing')
ifelse((is.na(x) | is.nan(x) | is.infinite(x)), TRUE,
ifelse(trimws(tolower(x)) %in% missing_strs, TRUE, FALSE))
}
missing_summary<-summarise_all(bankruptcy.clean, ~(sum(is_missing(.) / nrow(bankruptcy.clean))))
stack(sort(missing_summary, decreasing = TRUE))## values ind
## 1 0 Bankrupt
## 2 0 ROA.C..before.interest.and.depreciation.before.interest
## 3 0 ROA.A..before.interest.and...after.tax
## 4 0 ROA.B..before.interest.and.depreciation.after.tax
## 5 0 Operating.Gross.Margin
## 6 0 Realized.Sales.Gross.Margin
## 7 0 Operating.Profit.Rate
## 8 0 Pre.tax.net.Interest.Rate
## 9 0 After.tax.net.Interest.Rate
## 10 0 Non.industry.income.and.expenditure.revenue
## 11 0 Continuous.interest.rate..after.tax.
## 12 0 Operating.Expense.Rate
## 13 0 Research.and.development.expense.rate
## 14 0 Cash.flow.rate
## 15 0 Interest.bearing.debt.interest.rate
## 16 0 Tax.rate..A.
## 17 0 Net.Value.Per.Share..B.
## 18 0 Net.Value.Per.Share..A.
## 19 0 Net.Value.Per.Share..C.
## 20 0 Persistent.EPS.in.the.Last.Four.Seasons
## 21 0 Cash.Flow.Per.Share
## 22 0 Revenue.Per.Share..Yuan.Â..
## 23 0 Operating.Profit.Per.Share..Yuan.Â..
## 24 0 Per.Share.Net.profit.before.tax..Yuan.Â..
## 25 0 Realized.Sales.Gross.Profit.Growth.Rate
## 26 0 Operating.Profit.Growth.Rate
## 27 0 After.tax.Net.Profit.Growth.Rate
## 28 0 Regular.Net.Profit.Growth.Rate
## 29 0 Continuous.Net.Profit.Growth.Rate
## 30 0 Total.Asset.Growth.Rate
## 31 0 Net.Value.Growth.Rate
## 32 0 Total.Asset.Return.Growth.Rate.Ratio
## 33 0 Cash.Reinvestment..
## 34 0 Current.Ratio
## 35 0 Quick.Ratio
## 36 0 Interest.Expense.Ratio
## 37 0 Total.debt.Total.net.worth
## 38 0 Debt.ratio..
## 39 0 Net.worth.Assets
## 40 0 Long.term.fund.suitability.ratio..A.
## 41 0 Borrowing.dependency
## 42 0 Contingent.liabilities.Net.worth
## 43 0 Operating.profit.Paid.in.capital
## 44 0 Net.profit.before.tax.Paid.in.capital
## 45 0 Inventory.and.accounts.receivable.Net.value
## 46 0 Total.Asset.Turnover
## 47 0 Accounts.Receivable.Turnover
## 48 0 Average.Collection.Days
## 49 0 Inventory.Turnover.Rate..times.
## 50 0 Fixed.Assets.Turnover.Frequency
## 51 0 Net.Worth.Turnover.Rate..times.
## 52 0 Revenue.per.person
## 53 0 Operating.profit.per.person
## 54 0 Allocation.rate.per.person
## 55 0 Working.Capital.to.Total.Assets
## 56 0 Quick.Assets.Total.Assets
## 57 0 Current.Assets.Total.Assets
## 58 0 Cash.Total.Assets
## 59 0 Quick.Assets.Current.Liability
## 60 0 Cash.Current.Liability
## 61 0 Current.Liability.to.Assets
## 62 0 Operating.Funds.to.Liability
## 63 0 Inventory.Working.Capital
## 64 0 Inventory.Current.Liability
## 65 0 Current.Liabilities.Liability
## 66 0 Working.Capital.Equity
## 67 0 Current.Liabilities.Equity
## 68 0 Long.term.Liability.to.Current.Assets
## 69 0 Retained.Earnings.to.Total.Assets
## 70 0 Total.income.Total.expense
## 71 0 Total.expense.Assets
## 72 0 Current.Asset.Turnover.Rate
## 73 0 Quick.Asset.Turnover.Rate
## 74 0 Working.capitcal.Turnover.Rate
## 75 0 Cash.Turnover.Rate
## 76 0 Cash.Flow.to.Sales
## 77 0 Fixed.Assets.to.Assets
## 78 0 Current.Liability.to.Liability
## 79 0 Current.Liability.to.Equity
## 80 0 Equity.to.Long.term.Liability
## 81 0 Cash.Flow.to.Total.Assets
## 82 0 Cash.Flow.to.Liability
## 83 0 CFO.to.Assets
## 84 0 Cash.Flow.to.Equity
## 85 0 Current.Liability.to.Current.Assets
## 86 0 Liability.Assets.Flag
## 87 0 Net.Income.to.Total.Assets
## 88 0 Total.assets.to.GNP.price
## 89 0 No.credit.Interval
## 90 0 Gross.Profit.to.Sales
## 91 0 Net.Income.to.Stockholder.s.Equity
## 92 0 Liability.to.Equity
## 93 0 Degree.of.Financial.Leverage..DFL.
## 94 0 Interest.Coverage.Ratio..Interest.expense.to.EBIT.Prepare data
set.seed(1)
#divide into training and testing data
inTrain <- createDataPartition(bankruptcy.clean$Bankrupt, p = .75, list = F)#training set 75%
training_set <- bankruptcy.clean[ inTrain, ]
test_set <- bankruptcy.clean[-inTrain, ]Model Development
logit_1 <- glm(Bankrupt~., family = binomial,data = training_set)
summary(logit_1)##
## Call:
## glm(formula = Bankrupt ~ ., family = binomial, data = training_set)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -8.49 0.00 0.00 0.00 8.49
##
## Coefficients: (3 not defined because of singularities)
## Estimate Std. Error
## (Intercept) 2.171e+23 5.269e+15
## ROA.C..before.interest.and.depreciation.before.interest -5.858e+15 1.444e+08
## ROA.A..before.interest.and...after.tax 3.513e+16 1.133e+08
## ROA.B..before.interest.and.depreciation.after.tax -3.018e+15 1.799e+08
## Operating.Gross.Margin 1.528e+19 4.555e+11
## Realized.Sales.Gross.Margin -3.958e+16 1.861e+09
## Operating.Profit.Rate -1.891e+21 1.636e+13
## Pre.tax.net.Interest.Rate 1.576e+21 1.366e+13
## After.tax.net.Interest.Rate -5.871e+17 7.603e+09
## Non.industry.income.and.expenditure.revenue -9.025e+20 7.821e+12
## Continuous.interest.rate..after.tax. 2.630e+18 5.781e+09
## Operating.Expense.Rate 3.386e+14 3.270e+06
## Research.and.development.expense.rate 3.463e+14 3.796e+06
## Cash.flow.rate 2.346e+16 2.061e+08
## Interest.bearing.debt.interest.rate -1.033e+14 8.786e+06
## Tax.rate..A. 4.522e+14 7.761e+06
## Net.Value.Per.Share..B. -5.321e+16 6.952e+08
## Net.Value.Per.Share..A. 2.914e+16 2.127e+09
## Net.Value.Per.Share..C. 2.276e+16 2.012e+09
## Persistent.EPS.in.the.Last.Four.Seasons -2.289e+16 2.435e+08
## Cash.Flow.Per.Share 4.459e+15 1.104e+08
## Revenue.Per.Share..Yuan.Â.. -3.388e+16 2.124e+08
## Operating.Profit.Per.Share..Yuan.Â.. 3.364e+16 9.902e+08
## Per.Share.Net.profit.before.tax..Yuan.Â.. -1.457e+16 1.316e+08
## Realized.Sales.Gross.Profit.Growth.Rate -5.167e+17 6.937e+07
## Operating.Profit.Growth.Rate -2.881e+16 3.063e+08
## After.tax.Net.Profit.Growth.Rate -1.481e+19 6.835e+08
## Regular.Net.Profit.Growth.Rate 1.481e+19 6.773e+08
## Continuous.Net.Profit.Growth.Rate -2.164e+17 8.190e+07
## Total.Asset.Growth.Rate 9.233e+13 3.502e+06
## Net.Value.Growth.Rate 8.069e+16 1.051e+09
## Total.Asset.Return.Growth.Rate.Ratio -5.473e+15 1.296e+08
## Cash.Reinvestment.. 3.683e+15 8.877e+07
## Current.Ratio 4.065e+15 8.405e+07
## Quick.Ratio -3.829e+15 4.882e+07
## Interest.Expense.Ratio -4.530e+15 1.072e+08
## Total.debt.Total.net.worth -6.719e+15 7.104e+07
## Debt.ratio.. 2.409e+16 9.393e+07
## Net.worth.Assets NA NA
## Long.term.fund.suitability.ratio..A. -4.157e+15 3.782e+07
## Borrowing.dependency 2.030e+16 3.741e+08
## Contingent.liabilities.Net.worth 7.822e+15 2.609e+08
## Operating.profit.Paid.in.capital -9.573e+15 9.959e+08
## Net.profit.before.tax.Paid.in.capital 7.962e+15 2.715e+08
## Inventory.and.accounts.receivable.Net.value 4.701e+15 4.380e+08
## Total.Asset.Turnover 1.963e+15 2.510e+07
## Accounts.Receivable.Turnover -1.767e+16 4.323e+07
## Average.Collection.Days -1.001e+16 4.553e+07
## Inventory.Turnover.Rate..times. -8.249e+13 3.083e+06
## Fixed.Assets.Turnover.Frequency 3.059e+14 4.417e+06
## Net.Worth.Turnover.Rate..times. -1.566e+16 7.299e+07
## Revenue.per.person -5.536e+15 2.204e+08
## Operating.profit.per.person 4.120e+14 3.990e+07
## Allocation.rate.per.person -1.154e+15 3.441e+07
## Working.Capital.to.Total.Assets -2.875e+23 7.006e+15
## Quick.Assets.Total.Assets 9.928e+14 1.041e+07
## Current.Assets.Total.Assets 7.876e+22 1.919e+15
## Cash.Total.Assets -1.137e+15 1.195e+07
## Quick.Assets.Current.Liability 6.028e+14 6.747e+07
## Cash.Current.Liability 4.744e+14 1.754e+07
## Current.Liability.to.Assets -2.565e+23 6.250e+15
## Operating.Funds.to.Liability -5.418e+13 9.064e+07
## Inventory.Working.Capital -5.422e+15 8.654e+07
## Inventory.Current.Liability -1.336e+15 1.698e+07
## Current.Liabilities.Liability 2.174e+15 1.439e+07
## Working.Capital.Equity -5.537e+15 5.812e+08
## Current.Liabilities.Equity 1.505e+16 1.180e+09
## Long.term.Liability.to.Current.Assets -1.099e+14 1.480e+07
## Retained.Earnings.to.Total.Assets 4.201e+15 7.436e+07
## Total.income.Total.expense -4.829e+17 2.670e+09
## Total.expense.Assets -1.038e+16 6.998e+07
## Current.Asset.Turnover.Rate 1.668e+14 4.083e+06
## Quick.Asset.Turnover.Rate 2.523e+14 3.333e+06
## Working.capitcal.Turnover.Rate -2.684e+17 2.539e+09
## Cash.Turnover.Rate 7.230e+13 3.460e+06
## Cash.Flow.to.Sales 3.460e+17 8.844e+08
## Fixed.Assets.to.Assets 4.614e+15 7.496e+07
## Current.Liability.to.Liability NA NA
## Current.Liability.to.Equity NA NA
## Equity.to.Long.term.Liability 6.480e+15 3.305e+08
## Cash.Flow.to.Total.Assets -3.063e+15 4.368e+07
## Cash.Flow.to.Liability -4.277e+14 5.218e+07
## CFO.to.Assets -3.160e+15 4.391e+07
## Cash.Flow.to.Equity -2.578e+14 1.431e+08
## Current.Liability.to.Current.Assets -2.254e+15 5.720e+07
## Liability.Assets.Flag 5.350e+14 3.798e+07
## Net.Income.to.Total.Assets -4.559e+16 1.467e+08
## Total.assets.to.GNP.price -4.532e+14 2.451e+07
## No.credit.Interval 1.033e+15 7.557e+07
## Gross.Profit.to.Sales -1.521e+19 4.555e+11
## Net.Income.to.Stockholder.s.Equity 9.086e+14 2.304e+08
## Liability.to.Equity -4.593e+16 1.431e+09
## Degree.of.Financial.Leverage..DFL. 1.518e+15 5.346e+07
## Interest.Coverage.Ratio..Interest.expense.to.EBIT. -1.703e+15 8.026e+07
## z value Pr(>|z|)
## (Intercept) 4.121e+07 <2e-16 ***
## ROA.C..before.interest.and.depreciation.before.interest -4.056e+07 <2e-16 ***
## ROA.A..before.interest.and...after.tax 3.101e+08 <2e-16 ***
## ROA.B..before.interest.and.depreciation.after.tax -1.677e+07 <2e-16 ***
## Operating.Gross.Margin 3.355e+07 <2e-16 ***
## Realized.Sales.Gross.Margin -2.126e+07 <2e-16 ***
## Operating.Profit.Rate -1.156e+08 <2e-16 ***
## Pre.tax.net.Interest.Rate 1.153e+08 <2e-16 ***
## After.tax.net.Interest.Rate -7.722e+07 <2e-16 ***
## Non.industry.income.and.expenditure.revenue -1.154e+08 <2e-16 ***
## Continuous.interest.rate..after.tax. 4.549e+08 <2e-16 ***
## Operating.Expense.Rate 1.035e+08 <2e-16 ***
## Research.and.development.expense.rate 9.122e+07 <2e-16 ***
## Cash.flow.rate 1.138e+08 <2e-16 ***
## Interest.bearing.debt.interest.rate -1.176e+07 <2e-16 ***
## Tax.rate..A. 5.827e+07 <2e-16 ***
## Net.Value.Per.Share..B. -7.654e+07 <2e-16 ***
## Net.Value.Per.Share..A. 1.370e+07 <2e-16 ***
## Net.Value.Per.Share..C. 1.131e+07 <2e-16 ***
## Persistent.EPS.in.the.Last.Four.Seasons -9.400e+07 <2e-16 ***
## Cash.Flow.Per.Share 4.038e+07 <2e-16 ***
## Revenue.Per.Share..Yuan.Â.. -1.595e+08 <2e-16 ***
## Operating.Profit.Per.Share..Yuan.Â.. 3.397e+07 <2e-16 ***
## Per.Share.Net.profit.before.tax..Yuan.Â.. -1.107e+08 <2e-16 ***
## Realized.Sales.Gross.Profit.Growth.Rate -7.449e+09 <2e-16 ***
## Operating.Profit.Growth.Rate -9.405e+07 <2e-16 ***
## After.tax.Net.Profit.Growth.Rate -2.167e+10 <2e-16 ***
## Regular.Net.Profit.Growth.Rate 2.187e+10 <2e-16 ***
## Continuous.Net.Profit.Growth.Rate -2.643e+09 <2e-16 ***
## Total.Asset.Growth.Rate 2.636e+07 <2e-16 ***
## Net.Value.Growth.Rate 7.681e+07 <2e-16 ***
## Total.Asset.Return.Growth.Rate.Ratio -4.222e+07 <2e-16 ***
## Cash.Reinvestment.. 4.149e+07 <2e-16 ***
## Current.Ratio 4.837e+07 <2e-16 ***
## Quick.Ratio -7.843e+07 <2e-16 ***
## Interest.Expense.Ratio -4.228e+07 <2e-16 ***
## Total.debt.Total.net.worth -9.459e+07 <2e-16 ***
## Debt.ratio.. 2.565e+08 <2e-16 ***
## Net.worth.Assets NA NA
## Long.term.fund.suitability.ratio..A. -1.099e+08 <2e-16 ***
## Borrowing.dependency 5.426e+07 <2e-16 ***
## Contingent.liabilities.Net.worth 2.998e+07 <2e-16 ***
## Operating.profit.Paid.in.capital -9.612e+06 <2e-16 ***
## Net.profit.before.tax.Paid.in.capital 2.933e+07 <2e-16 ***
## Inventory.and.accounts.receivable.Net.value 1.073e+07 <2e-16 ***
## Total.Asset.Turnover 7.824e+07 <2e-16 ***
## Accounts.Receivable.Turnover -4.087e+08 <2e-16 ***
## Average.Collection.Days -2.197e+08 <2e-16 ***
## Inventory.Turnover.Rate..times. -2.675e+07 <2e-16 ***
## Fixed.Assets.Turnover.Frequency 6.926e+07 <2e-16 ***
## Net.Worth.Turnover.Rate..times. -2.146e+08 <2e-16 ***
## Revenue.per.person -2.512e+07 <2e-16 ***
## Operating.profit.per.person 1.033e+07 <2e-16 ***
## Allocation.rate.per.person -3.354e+07 <2e-16 ***
## Working.Capital.to.Total.Assets -4.104e+07 <2e-16 ***
## Quick.Assets.Total.Assets 9.535e+07 <2e-16 ***
## Current.Assets.Total.Assets 4.104e+07 <2e-16 ***
## Cash.Total.Assets -9.513e+07 <2e-16 ***
## Quick.Assets.Current.Liability 8.934e+06 <2e-16 ***
## Cash.Current.Liability 2.706e+07 <2e-16 ***
## Current.Liability.to.Assets -4.104e+07 <2e-16 ***
## Operating.Funds.to.Liability -5.978e+05 <2e-16 ***
## Inventory.Working.Capital -6.265e+07 <2e-16 ***
## Inventory.Current.Liability -7.868e+07 <2e-16 ***
## Current.Liabilities.Liability 1.511e+08 <2e-16 ***
## Working.Capital.Equity -9.527e+06 <2e-16 ***
## Current.Liabilities.Equity 1.275e+07 <2e-16 ***
## Long.term.Liability.to.Current.Assets -7.426e+06 <2e-16 ***
## Retained.Earnings.to.Total.Assets 5.649e+07 <2e-16 ***
## Total.income.Total.expense -1.809e+08 <2e-16 ***
## Total.expense.Assets -1.483e+08 <2e-16 ***
## Current.Asset.Turnover.Rate 4.086e+07 <2e-16 ***
## Quick.Asset.Turnover.Rate 7.570e+07 <2e-16 ***
## Working.capitcal.Turnover.Rate -1.057e+08 <2e-16 ***
## Cash.Turnover.Rate 2.090e+07 <2e-16 ***
## Cash.Flow.to.Sales 3.913e+08 <2e-16 ***
## Fixed.Assets.to.Assets 6.156e+07 <2e-16 ***
## Current.Liability.to.Liability NA NA
## Current.Liability.to.Equity NA NA
## Equity.to.Long.term.Liability 1.961e+07 <2e-16 ***
## Cash.Flow.to.Total.Assets -7.014e+07 <2e-16 ***
## Cash.Flow.to.Liability -8.198e+06 <2e-16 ***
## CFO.to.Assets -7.196e+07 <2e-16 ***
## Cash.Flow.to.Equity -1.802e+06 <2e-16 ***
## Current.Liability.to.Current.Assets -3.940e+07 <2e-16 ***
## Liability.Assets.Flag 1.409e+07 <2e-16 ***
## Net.Income.to.Total.Assets -3.108e+08 <2e-16 ***
## Total.assets.to.GNP.price -1.849e+07 <2e-16 ***
## No.credit.Interval 1.367e+07 <2e-16 ***
## Gross.Profit.to.Sales -3.338e+07 <2e-16 ***
## Net.Income.to.Stockholder.s.Equity 3.944e+06 <2e-16 ***
## Liability.to.Equity -3.210e+07 <2e-16 ***
## Degree.of.Financial.Leverage..DFL. 2.840e+07 <2e-16 ***
## Interest.Coverage.Ratio..Interest.expense.to.EBIT. -2.122e+07 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 1457.8 on 5114 degrees of freedom
## Residual deviance: 17877.7 on 5024 degrees of freedom
## AIC: 18060
##
## Number of Fisher Scoring iterations: 17Now for efficiency and because we had run a logit model earlier in the semester, let’s build an AIC model. AIC Stands for Akaike Information Criterion, and ultimately is an estimator of in-sample prediction error and is similar to the adjusted R-squared measures we see in our regression output summaries. It will give us the features with the lowest score AIC. Lower scores can indicate a more parsimonious model, relative to a model fit with a higher AIC. It can therefore give an indication of the relative quality of statistical models for a given set of data.
#logit_3 <- stepAIC(logit_1, direction = "backward")
logit_3 <- readRDS("final_aic.rds")
summary(logit_3)##
## Call:
## glm(formula = Bankrupt ~ ROA.C..before.interest.and.depreciation.before.interest +
## Pre.tax.net.Interest.Rate + Cash.flow.rate + Net.Value.Per.Share..B. +
## Net.Value.Per.Share..C. + Persistent.EPS.in.the.Last.Four.Seasons +
## Operating.Profit.Growth.Rate + Cash.Reinvestment.. + Debt.ratio.. +
## Borrowing.dependency + Operating.profit.Paid.in.capital +
## Total.Asset.Turnover + Accounts.Receivable.Turnover + Revenue.per.person +
## Operating.profit.per.person + Quick.Assets.Total.Assets +
## Cash.Total.Assets + Cash.Current.Liability + Current.Liability.to.Assets +
## Current.Liabilities.Liability + Working.Capital.Equity +
## Current.Liabilities.Equity + Cash.Turnover.Rate + Fixed.Assets.to.Assets +
## Equity.to.Long.term.Liability + Cash.Flow.to.Total.Assets +
## Liability.Assets.Flag + Liability.to.Equity, family = binomial,
## data = training_set)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -3.3919 -0.1668 -0.0798 -0.0356 3.7216
##
## Coefficients:
## Estimate Std. Error
## (Intercept) 14.0297 92.3977
## ROA.C..before.interest.and.depreciation.before.interest -10.1596 3.6290
## Pre.tax.net.Interest.Rate 98.2074 118.2243
## Cash.flow.rate -19.6468 9.3738
## Net.Value.Per.Share..B. -151.4631 62.5933
## Net.Value.Per.Share..C. 144.3462 61.1721
## Persistent.EPS.in.the.Last.Four.Seasons -45.9139 11.1659
## Operating.Profit.Growth.Rate -61.6672 20.4207
## Cash.Reinvestment.. 13.1558 4.4724
## Debt.ratio.. 42.0130 10.1384
## Borrowing.dependency 49.2731 18.8769
## Operating.profit.Paid.in.capital 37.7966 12.1538
## Total.Asset.Turnover -5.8001 1.7736
## Accounts.Receivable.Turnover -451.0242 17325.6193
## Revenue.per.person 99.1115 22566.5788
## Operating.profit.per.person -9.6809 6.5689
## Quick.Assets.Total.Assets 1.1761 0.7939
## Cash.Total.Assets -6.4419 1.9548
## Cash.Current.Liability 1.1019 0.7086
## Current.Liability.to.Assets -26.8032 12.1972
## Current.Liabilities.Liability 4.3072 1.9735
## Working.Capital.Equity -34.6268 12.9214
## Current.Liabilities.Equity 138.7132 65.6824
## Cash.Turnover.Rate -1.1436 0.4224
## Fixed.Assets.to.Assets 22.6078 17720.7527
## Equity.to.Long.term.Liability 38.1454 18.5837
## Cash.Flow.to.Total.Assets -7.2718 2.5117
## Liability.Assets.Flag -6.5825 2.2724
## Liability.to.Equity -240.9508 83.2862
## z value Pr(>|z|)
## (Intercept) 0.152 0.879313
## ROA.C..before.interest.and.depreciation.before.interest -2.800 0.005117 **
## Pre.tax.net.Interest.Rate 0.831 0.406151
## Cash.flow.rate -2.096 0.036088 *
## Net.Value.Per.Share..B. -2.420 0.015529 *
## Net.Value.Per.Share..C. 2.360 0.018291 *
## Persistent.EPS.in.the.Last.Four.Seasons -4.112 3.92e-05 ***
## Operating.Profit.Growth.Rate -3.020 0.002529 **
## Cash.Reinvestment.. 2.942 0.003266 **
## Debt.ratio.. 4.144 3.41e-05 ***
## Borrowing.dependency 2.610 0.009048 **
## Operating.profit.Paid.in.capital 3.110 0.001872 **
## Total.Asset.Turnover -3.270 0.001074 **
## Accounts.Receivable.Turnover -0.026 0.979232
## Revenue.per.person 0.004 0.996496
## Operating.profit.per.person -1.474 0.140551
## Quick.Assets.Total.Assets 1.481 0.138502
## Cash.Total.Assets -3.295 0.000982 ***
## Cash.Current.Liability 1.555 0.119928
## Current.Liability.to.Assets -2.197 0.027985 *
## Current.Liabilities.Liability 2.183 0.029072 *
## Working.Capital.Equity -2.680 0.007366 **
## Current.Liabilities.Equity 2.112 0.034697 *
## Cash.Turnover.Rate -2.707 0.006780 **
## Fixed.Assets.to.Assets 0.001 0.998982
## Equity.to.Long.term.Liability 2.053 0.040109 *
## Cash.Flow.to.Total.Assets -2.895 0.003790 **
## Liability.Assets.Flag -2.897 0.003771 **
## Liability.to.Equity -2.893 0.003815 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 1457.84 on 5114 degrees of freedom
## Residual deviance: 821.24 on 5086 degrees of freedom
## AIC: 879.24
##
## Number of Fisher Scoring iterations: 19We also know this is a highly class imbalanced dataset, however, with our bankrupted companies making up the minority class. We use the care upSample package to create an upsampled training dataset, and create a logit model and AIC model with that dataset as well:
set.seed(1)
training_set$Bankrupt <- as.factor(training_set$Bankrupt)
table(training_set$Bankrupt)##
## NO YES
## 4950 165up_train <- upSample(x = training_set[, -1],
y=training_set$Bankrupt,
yname = 'Bankrupt')
table(up_train$Bankrupt)##
## NO YES
## 4950 4950logit_1_up <- glm(Bankrupt~., family = binomial,data = up_train)
summary(logit_1_up)##
## Call:
## glm(formula = Bankrupt ~ ., family = binomial, data = up_train)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -8.49 0.00 0.00 0.00 8.49
##
## Coefficients: (3 not defined because of singularities)
## Estimate Std. Error
## (Intercept) 1.982e+23 4.063e+15
## ROA.C..before.interest.and.depreciation.before.interest -7.851e+15 8.785e+07
## ROA.A..before.interest.and...after.tax 1.520e+16 7.194e+07
## ROA.B..before.interest.and.depreciation.after.tax 4.839e+15 1.178e+08
## Operating.Gross.Margin 3.993e+19 3.409e+11
## Realized.Sales.Gross.Margin -4.822e+16 1.714e+09
## Operating.Profit.Rate -8.496e+20 1.285e+13
## Pre.tax.net.Interest.Rate 7.097e+20 1.073e+13
## After.tax.net.Interest.Rate 3.352e+17 6.521e+09
## Non.industry.income.and.expenditure.revenue -4.061e+20 6.144e+12
## Continuous.interest.rate..after.tax. -4.580e+17 3.847e+09
## Operating.Expense.Rate 1.971e+14 2.424e+06
## Research.and.development.expense.rate 2.914e+14 3.017e+06
## Cash.flow.rate -3.218e+16 1.777e+08
## Interest.bearing.debt.interest.rate -3.097e+14 8.712e+06
## Tax.rate..A. 5.490e+14 6.174e+06
## Net.Value.Per.Share..B. -6.313e+16 5.494e+08
## Net.Value.Per.Share..A. 5.622e+16 1.177e+09
## Net.Value.Per.Share..C. 7.006e+15 1.033e+09
## Persistent.EPS.in.the.Last.Four.Seasons -2.612e+16 1.855e+08
## Cash.Flow.Per.Share 1.932e+15 8.799e+07
## Revenue.Per.Share..Yuan.Â.. 1.098e+16 1.902e+08
## Operating.Profit.Per.Share..Yuan.Â.. -1.186e+17 7.722e+08
## Per.Share.Net.profit.before.tax..Yuan.Â.. 6.876e+15 1.242e+08
## Realized.Sales.Gross.Profit.Growth.Rate 3.370e+13 6.573e+07
## Operating.Profit.Growth.Rate -1.609e+16 2.223e+08
## After.tax.Net.Profit.Growth.Rate 1.902e+16 5.613e+08
## Regular.Net.Profit.Growth.Rate -1.084e+16 5.594e+08
## Continuous.Net.Profit.Growth.Rate 3.453e+14 7.712e+07
## Total.Asset.Growth.Rate 4.380e+14 2.861e+06
## Net.Value.Growth.Rate 2.163e+16 7.291e+08
## Total.Asset.Return.Growth.Rate.Ratio -1.686e+15 1.151e+08
## Cash.Reinvestment.. -2.318e+14 3.535e+07
## Current.Ratio 7.817e+14 7.398e+07
## Quick.Ratio -5.852e+15 4.799e+07
## Interest.Expense.Ratio -3.280e+15 5.835e+07
## Total.debt.Total.net.worth 2.232e+15 6.727e+07
## Debt.ratio.. 2.716e+15 5.344e+07
## Net.worth.Assets NA NA
## Long.term.fund.suitability.ratio..A. 1.426e+14 3.158e+07
## Borrowing.dependency 8.281e+15 1.612e+08
## Contingent.liabilities.Net.worth -5.472e+15 1.139e+08
## Operating.profit.Paid.in.capital 1.314e+17 7.806e+08
## Net.profit.before.tax.Paid.in.capital -5.892e+13 2.237e+08
## Inventory.and.accounts.receivable.Net.value -2.864e+16 2.419e+08
## Total.Asset.Turnover 2.046e+15 1.899e+07
## Accounts.Receivable.Turnover -8.869e+15 4.141e+07
## Average.Collection.Days -6.795e+15 4.515e+07
## Inventory.Turnover.Rate..times. -3.851e+14 2.359e+06
## Fixed.Assets.Turnover.Frequency 4.166e+14 2.975e+06
## Net.Worth.Turnover.Rate..times. -8.812e+15 5.337e+07
## Revenue.per.person 1.494e+16 1.765e+08
## Operating.profit.per.person -1.208e+15 3.370e+07
## Allocation.rate.per.person -1.799e+15 3.362e+07
## Working.Capital.to.Total.Assets -2.630e+23 5.404e+15
## Quick.Assets.Total.Assets -3.100e+14 7.384e+06
## Current.Assets.Total.Assets 7.204e+22 1.480e+15
## Cash.Total.Assets -1.835e+15 9.701e+06
## Quick.Assets.Current.Liability 6.034e+14 6.733e+07
## Cash.Current.Liability 1.082e+15 7.181e+06
## Current.Liability.to.Assets -2.346e+23 4.820e+15
## Operating.Funds.to.Liability 6.624e+15 7.656e+07
## Inventory.Working.Capital -7.562e+15 8.098e+07
## Inventory.Current.Liability 4.361e+14 1.402e+07
## Current.Liabilities.Liability -1.222e+13 1.001e+07
## Working.Capital.Equity 2.526e+16 3.102e+08
## Current.Liabilities.Equity 8.641e+16 5.935e+08
## Long.term.Liability.to.Current.Assets -9.112e+14 1.001e+07
## Retained.Earnings.to.Total.Assets -2.495e+15 5.661e+07
## Total.income.Total.expense -1.583e+17 2.477e+09
## Total.expense.Assets -8.844e+14 4.230e+07
## Current.Asset.Turnover.Rate 8.103e+13 2.977e+06
## Quick.Asset.Turnover.Rate -7.042e+13 2.480e+06
## Working.capitcal.Turnover.Rate -8.844e+16 2.221e+09
## Cash.Turnover.Rate -4.868e+14 2.796e+06
## Cash.Flow.to.Sales -1.750e+15 8.031e+08
## Fixed.Assets.to.Assets 4.939e+15 2.998e+07
## Current.Liability.to.Liability NA NA
## Current.Liability.to.Equity NA NA
## Equity.to.Long.term.Liability 1.671e+16 1.885e+08
## Cash.Flow.to.Total.Assets 2.955e+15 3.097e+07
## Cash.Flow.to.Liability -4.745e+15 4.205e+07
## CFO.to.Assets 1.437e+15 2.836e+07
## Cash.Flow.to.Equity -6.762e+15 9.215e+07
## Current.Liability.to.Current.Assets 2.052e+15 3.442e+07
## Liability.Assets.Flag 7.706e+14 1.459e+07
## Net.Income.to.Total.Assets -2.573e+16 7.801e+07
## Total.assets.to.GNP.price 1.067e+15 1.308e+07
## No.credit.Interval -4.834e+15 7.259e+07
## Gross.Profit.to.Sales -3.989e+19 3.409e+11
## Net.Income.to.Stockholder.s.Equity 1.059e+16 7.970e+07
## Liability.to.Equity -6.038e+16 8.285e+08
## Degree.of.Financial.Leverage..DFL. 1.480e+14 5.283e+07
## Interest.Coverage.Ratio..Interest.expense.to.EBIT. 3.935e+15 7.288e+07
## z value Pr(>|z|)
## (Intercept) 48774985 <2e-16 ***
## ROA.C..before.interest.and.depreciation.before.interest -89363994 <2e-16 ***
## ROA.A..before.interest.and...after.tax 211267754 <2e-16 ***
## ROA.B..before.interest.and.depreciation.after.tax 41072030 <2e-16 ***
## Operating.Gross.Margin 117155887 <2e-16 ***
## Realized.Sales.Gross.Margin -28128004 <2e-16 ***
## Operating.Profit.Rate -66106236 <2e-16 ***
## Pre.tax.net.Interest.Rate 66119344 <2e-16 ***
## After.tax.net.Interest.Rate 51403458 <2e-16 ***
## Non.industry.income.and.expenditure.revenue -66103101 <2e-16 ***
## Continuous.interest.rate..after.tax. -119062556 <2e-16 ***
## Operating.Expense.Rate 81314459 <2e-16 ***
## Research.and.development.expense.rate 96582471 <2e-16 ***
## Cash.flow.rate -181139773 <2e-16 ***
## Interest.bearing.debt.interest.rate -35549670 <2e-16 ***
## Tax.rate..A. 88914620 <2e-16 ***
## Net.Value.Per.Share..B. -114913491 <2e-16 ***
## Net.Value.Per.Share..A. 47752353 <2e-16 ***
## Net.Value.Per.Share..C. 6778975 <2e-16 ***
## Persistent.EPS.in.the.Last.Four.Seasons -140836983 <2e-16 ***
## Cash.Flow.Per.Share 21953193 <2e-16 ***
## Revenue.Per.Share..Yuan.Â.. 57707717 <2e-16 ***
## Operating.Profit.Per.Share..Yuan.Â.. -153638273 <2e-16 ***
## Per.Share.Net.profit.before.tax..Yuan.Â.. 55364741 <2e-16 ***
## Realized.Sales.Gross.Profit.Growth.Rate 512743 <2e-16 ***
## Operating.Profit.Growth.Rate -72378464 <2e-16 ***
## After.tax.Net.Profit.Growth.Rate 33894274 <2e-16 ***
## Regular.Net.Profit.Growth.Rate -19376351 <2e-16 ***
## Continuous.Net.Profit.Growth.Rate 4477255 <2e-16 ***
## Total.Asset.Growth.Rate 153109118 <2e-16 ***
## Net.Value.Growth.Rate 29662706 <2e-16 ***
## Total.Asset.Return.Growth.Rate.Ratio -14648422 <2e-16 ***
## Cash.Reinvestment.. -6558039 <2e-16 ***
## Current.Ratio 10565354 <2e-16 ***
## Quick.Ratio -121941091 <2e-16 ***
## Interest.Expense.Ratio -56220547 <2e-16 ***
## Total.debt.Total.net.worth 33187318 <2e-16 ***
## Debt.ratio.. 50823643 <2e-16 ***
## Net.worth.Assets NA NA
## Long.term.fund.suitability.ratio..A. 4516366 <2e-16 ***
## Borrowing.dependency 51376723 <2e-16 ***
## Contingent.liabilities.Net.worth -48051884 <2e-16 ***
## Operating.profit.Paid.in.capital 168335637 <2e-16 ***
## Net.profit.before.tax.Paid.in.capital -263452 <2e-16 ***
## Inventory.and.accounts.receivable.Net.value -118372477 <2e-16 ***
## Total.Asset.Turnover 107732110 <2e-16 ***
## Accounts.Receivable.Turnover -214201845 <2e-16 ***
## Average.Collection.Days -150511256 <2e-16 ***
## Inventory.Turnover.Rate..times. -163282993 <2e-16 ***
## Fixed.Assets.Turnover.Frequency 140029776 <2e-16 ***
## Net.Worth.Turnover.Rate..times. -165129573 <2e-16 ***
## Revenue.per.person 84641756 <2e-16 ***
## Operating.profit.per.person -35840179 <2e-16 ***
## Allocation.rate.per.person -53525092 <2e-16 ***
## Working.Capital.to.Total.Assets -48676155 <2e-16 ***
## Quick.Assets.Total.Assets -41984969 <2e-16 ***
## Current.Assets.Total.Assets 48676154 <2e-16 ***
## Cash.Total.Assets -189160341 <2e-16 ***
## Quick.Assets.Current.Liability 8961247 <2e-16 ***
## Cash.Current.Liability 150686382 <2e-16 ***
## Current.Liability.to.Assets -48676154 <2e-16 ***
## Operating.Funds.to.Liability 86530516 <2e-16 ***
## Inventory.Working.Capital -93382649 <2e-16 ***
## Inventory.Current.Liability 31109860 <2e-16 ***
## Current.Liabilities.Liability -1221326 <2e-16 ***
## Working.Capital.Equity 81423100 <2e-16 ***
## Current.Liabilities.Equity 145585258 <2e-16 ***
## Long.term.Liability.to.Current.Assets -91024388 <2e-16 ***
## Retained.Earnings.to.Total.Assets -44067120 <2e-16 ***
## Total.income.Total.expense -63892634 <2e-16 ***
## Total.expense.Assets -20906800 <2e-16 ***
## Current.Asset.Turnover.Rate 27214347 <2e-16 ***
## Quick.Asset.Turnover.Rate -28390980 <2e-16 ***
## Working.capitcal.Turnover.Rate -39822121 <2e-16 ***
## Cash.Turnover.Rate -174088436 <2e-16 ***
## Cash.Flow.to.Sales -2179647 <2e-16 ***
## Fixed.Assets.to.Assets 164715310 <2e-16 ***
## Current.Liability.to.Liability NA NA
## Current.Liability.to.Equity NA NA
## Equity.to.Long.term.Liability 88688438 <2e-16 ***
## Cash.Flow.to.Total.Assets 95431608 <2e-16 ***
## Cash.Flow.to.Liability -112820913 <2e-16 ***
## CFO.to.Assets 50672600 <2e-16 ***
## Cash.Flow.to.Equity -73378295 <2e-16 ***
## Current.Liability.to.Current.Assets 59616788 <2e-16 ***
## Liability.Assets.Flag 52801457 <2e-16 ***
## Net.Income.to.Total.Assets -329823548 <2e-16 ***
## Total.assets.to.GNP.price 81537944 <2e-16 ***
## No.credit.Interval -66593370 <2e-16 ***
## Gross.Profit.to.Sales -117001064 <2e-16 ***
## Net.Income.to.Stockholder.s.Equity 132864319 <2e-16 ***
## Liability.to.Equity -72874766 <2e-16 ***
## Degree.of.Financial.Leverage..DFL. 2800766 <2e-16 ***
## Interest.Coverage.Ratio..Interest.expense.to.EBIT. 53990995 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 13724 on 9899 degrees of freedom
## Residual deviance: 116709 on 9809 degrees of freedom
## AIC: 116891
##
## Number of Fisher Scoring iterations: 25#logit_3_up <- stepAIC(logit_1_up, direction = "backward")
logit_3_up <- readRDS("final_aic_up.rds")
summary(logit_3_up)##
## Call:
## glm(formula = Bankrupt ~ ROA.C..before.interest.and.depreciation.before.interest +
## ROA.A..before.interest.and...after.tax + ROA.B..before.interest.and.depreciation.after.tax +
## Operating.Gross.Margin + Realized.Sales.Gross.Margin + Operating.Profit.Rate +
## Pre.tax.net.Interest.Rate + After.tax.net.Interest.Rate +
## Non.industry.income.and.expenditure.revenue + Continuous.interest.rate..after.tax. +
## Operating.Expense.Rate + Research.and.development.expense.rate +
## Cash.flow.rate + Interest.bearing.debt.interest.rate + Tax.rate..A. +
## Net.Value.Per.Share..B. + Net.Value.Per.Share..A. + Net.Value.Per.Share..C. +
## Persistent.EPS.in.the.Last.Four.Seasons + Cash.Flow.Per.Share +
## Revenue.Per.Share..Yuan.Â.. + Operating.Profit.Per.Share..Yuan.Â.. +
## Per.Share.Net.profit.before.tax..Yuan.Â.. + Realized.Sales.Gross.Profit.Growth.Rate +
## Operating.Profit.Growth.Rate + After.tax.Net.Profit.Growth.Rate +
## Regular.Net.Profit.Growth.Rate + Continuous.Net.Profit.Growth.Rate +
## Total.Asset.Growth.Rate + Net.Value.Growth.Rate + Total.Asset.Return.Growth.Rate.Ratio +
## Cash.Reinvestment.. + Current.Ratio + Quick.Ratio + Interest.Expense.Ratio +
## Total.debt.Total.net.worth + Debt.ratio.. + Net.worth.Assets +
## Long.term.fund.suitability.ratio..A. + Borrowing.dependency +
## Contingent.liabilities.Net.worth + Operating.profit.Paid.in.capital +
## Net.profit.before.tax.Paid.in.capital + Inventory.and.accounts.receivable.Net.value +
## Total.Asset.Turnover + Accounts.Receivable.Turnover + Average.Collection.Days +
## Inventory.Turnover.Rate..times. + Fixed.Assets.Turnover.Frequency +
## Net.Worth.Turnover.Rate..times. + Revenue.per.person + Operating.profit.per.person +
## Allocation.rate.per.person + Working.Capital.to.Total.Assets +
## Quick.Assets.Total.Assets + Current.Assets.Total.Assets +
## Cash.Total.Assets + Quick.Assets.Current.Liability + Cash.Current.Liability +
## Current.Liability.to.Assets + Operating.Funds.to.Liability +
## Inventory.Working.Capital + Inventory.Current.Liability +
## Current.Liabilities.Liability + Working.Capital.Equity +
## Current.Liabilities.Equity + Long.term.Liability.to.Current.Assets +
## Retained.Earnings.to.Total.Assets + Total.income.Total.expense +
## Total.expense.Assets + Current.Asset.Turnover.Rate + Quick.Asset.Turnover.Rate +
## Working.capitcal.Turnover.Rate + Cash.Turnover.Rate + Cash.Flow.to.Sales +
## Fixed.Assets.to.Assets + Current.Liability.to.Liability +
## Current.Liability.to.Equity + Equity.to.Long.term.Liability +
## Cash.Flow.to.Total.Assets + Cash.Flow.to.Liability + CFO.to.Assets +
## Cash.Flow.to.Equity + Net.Income.to.Total.Assets + No.credit.Interval +
## Gross.Profit.to.Sales + Net.Income.to.Stockholder.s.Equity +
## Liability.to.Equity + Degree.of.Financial.Leverage..DFL. +
## Interest.Coverage.Ratio..Interest.expense.to.EBIT., family = binomial,
## data = up_train)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -8.49 0.00 0.00 0.00 8.49
##
## Coefficients: (3 not defined because of singularities)
## Estimate Std. Error
## (Intercept) 5.556e+21 4.054e+15
## ROA.C..before.interest.and.depreciation.before.interest -9.409e+15 7.644e+07
## ROA.A..before.interest.and...after.tax 1.755e+16 7.059e+07
## ROA.B..before.interest.and.depreciation.after.tax 4.287e+15 1.070e+08
## Operating.Gross.Margin 1.016e+19 3.406e+11
## Realized.Sales.Gross.Margin 9.469e+16 1.708e+09
## Operating.Profit.Rate -1.334e+21 1.280e+13
## Pre.tax.net.Interest.Rate 1.114e+21 1.069e+13
## After.tax.net.Interest.Rate 3.698e+17 6.483e+09
## Non.industry.income.and.expenditure.revenue -6.376e+20 6.121e+12
## Continuous.interest.rate..after.tax. -2.907e+17 3.727e+09
## Operating.Expense.Rate 1.410e+14 2.422e+06
## Research.and.development.expense.rate 2.505e+14 2.998e+06
## Cash.flow.rate -3.195e+16 1.759e+08
## Interest.bearing.debt.interest.rate -6.294e+14 8.704e+06
## Tax.rate..A. -7.492e+13 6.142e+06
## Net.Value.Per.Share..B. -7.869e+16 5.478e+08
## Net.Value.Per.Share..A. 1.129e+17 1.176e+09
## Net.Value.Per.Share..C. -3.475e+16 1.029e+09
## Persistent.EPS.in.the.Last.Four.Seasons -2.230e+16 1.791e+08
## Cash.Flow.Per.Share 4.972e+15 8.661e+07
## Revenue.Per.Share..Yuan.Â.. 1.437e+16 1.893e+08
## Operating.Profit.Per.Share..Yuan.Â.. -9.569e+16 7.694e+08
## Per.Share.Net.profit.before.tax..Yuan.Â.. 3.861e+15 1.241e+08
## Realized.Sales.Gross.Profit.Growth.Rate 3.028e+15 6.571e+07
## Operating.Profit.Growth.Rate -2.561e+16 2.003e+08
## After.tax.Net.Profit.Growth.Rate -6.637e+14 5.606e+08
## Regular.Net.Profit.Growth.Rate 5.053e+15 5.589e+08
## Continuous.Net.Profit.Growth.Rate 1.375e+15 7.709e+07
## Total.Asset.Growth.Rate 4.026e+14 2.856e+06
## Net.Value.Growth.Rate 2.445e+16 6.837e+08
## Total.Asset.Return.Growth.Rate.Ratio -2.665e+15 1.150e+08
## Cash.Reinvestment.. 3.586e+15 3.448e+07
## Current.Ratio -2.672e+15 6.767e+07
## Quick.Ratio -7.515e+15 4.798e+07
## Interest.Expense.Ratio -6.897e+15 5.834e+07
## Total.debt.Total.net.worth 2.545e+15 6.707e+07
## Debt.ratio.. 7.779e+15 4.828e+07
## Net.worth.Assets NA NA
## Long.term.fund.suitability.ratio..A. 1.048e+15 3.144e+07
## Borrowing.dependency 1.003e+16 1.570e+08
## Contingent.liabilities.Net.worth -2.639e+15 1.114e+08
## Operating.profit.Paid.in.capital 1.058e+17 7.778e+08
## Net.profit.before.tax.Paid.in.capital -3.441e+14 2.209e+08
## Inventory.and.accounts.receivable.Net.value -3.466e+16 2.233e+08
## Total.Asset.Turnover 2.407e+13 1.796e+07
## Accounts.Receivable.Turnover -1.002e+16 4.135e+07
## Average.Collection.Days -6.546e+15 4.514e+07
## Inventory.Turnover.Rate..times. -4.658e+14 2.355e+06
## Fixed.Assets.Turnover.Frequency 4.406e+14 2.968e+06
## Net.Worth.Turnover.Rate..times. -9.015e+15 4.924e+07
## Revenue.per.person 1.769e+16 1.758e+08
## Operating.profit.per.person -2.007e+15 3.364e+07
## Allocation.rate.per.person -2.119e+15 3.361e+07
## Working.Capital.to.Total.Assets -6.541e+21 5.391e+15
## Quick.Assets.Total.Assets 1.803e+14 7.247e+06
## Current.Assets.Total.Assets 1.792e+21 1.477e+15
## Cash.Total.Assets -1.315e+15 9.499e+06
## Quick.Assets.Current.Liability -3.656e+15 6.733e+07
## Cash.Current.Liability 1.717e+14 7.141e+06
## Current.Liability.to.Assets -5.834e+21 4.809e+15
## Operating.Funds.to.Liability 8.347e+15 7.627e+07
## Inventory.Working.Capital 2.372e+15 8.095e+07
## Inventory.Current.Liability -3.275e+14 1.127e+07
## Current.Liabilities.Liability 9.375e+14 9.620e+06
## Working.Capital.Equity 3.641e+16 2.799e+08
## Current.Liabilities.Equity 6.914e+16 5.907e+08
## Long.term.Liability.to.Current.Assets -1.576e+14 1.000e+07
## Retained.Earnings.to.Total.Assets 3.235e+15 5.463e+07
## Total.income.Total.expense 1.898e+15 2.421e+09
## Total.expense.Assets 4.729e+15 4.134e+07
## Current.Asset.Turnover.Rate 4.067e+14 2.966e+06
## Quick.Asset.Turnover.Rate -1.451e+14 2.473e+06
## Working.capitcal.Turnover.Rate -1.309e+17 2.202e+09
## Cash.Turnover.Rate -4.126e+14 2.789e+06
## Cash.Flow.to.Sales 2.593e+15 7.986e+08
## Fixed.Assets.to.Assets 4.838e+15 2.988e+07
## Current.Liability.to.Liability NA NA
## Current.Liability.to.Equity NA NA
## Equity.to.Long.term.Liability 9.807e+15 1.853e+08
## Cash.Flow.to.Total.Assets 4.061e+15 3.079e+07
## Cash.Flow.to.Liability -5.635e+15 4.184e+07
## CFO.to.Assets -7.739e+14 2.813e+07
## Cash.Flow.to.Equity -1.252e+16 9.140e+07
## Net.Income.to.Total.Assets -2.611e+16 7.429e+07
## No.credit.Interval -6.865e+15 7.259e+07
## Gross.Profit.to.Sales -1.026e+19 3.407e+11
## Net.Income.to.Stockholder.s.Equity 6.210e+15 7.584e+07
## Liability.to.Equity -3.524e+16 7.877e+08
## Degree.of.Financial.Leverage..DFL. -3.191e+15 5.277e+07
## Interest.Coverage.Ratio..Interest.expense.to.EBIT. -5.001e+14 7.286e+07
## z value Pr(>|z|)
## (Intercept) 1370611 <2e-16 ***
## ROA.C..before.interest.and.depreciation.before.interest -123088945 <2e-16 ***
## ROA.A..before.interest.and...after.tax 248614035 <2e-16 ***
## ROA.B..before.interest.and.depreciation.after.tax 40050839 <2e-16 ***
## Operating.Gross.Margin 29817538 <2e-16 ***
## Realized.Sales.Gross.Margin 55428482 <2e-16 ***
## Operating.Profit.Rate -104179772 <2e-16 ***
## Pre.tax.net.Interest.Rate 104176294 <2e-16 ***
## After.tax.net.Interest.Rate 57038948 <2e-16 ***
## Non.industry.income.and.expenditure.revenue -104171378 <2e-16 ***
## Continuous.interest.rate..after.tax. -78000272 <2e-16 ***
## Operating.Expense.Rate 58218147 <2e-16 ***
## Research.and.development.expense.rate 83534305 <2e-16 ***
## Cash.flow.rate -181641470 <2e-16 ***
## Interest.bearing.debt.interest.rate -72305299 <2e-16 ***
## Tax.rate..A. -12197460 <2e-16 ***
## Net.Value.Per.Share..B. -143651240 <2e-16 ***
## Net.Value.Per.Share..A. 96001928 <2e-16 ***
## Net.Value.Per.Share..C. -33782701 <2e-16 ***
## Persistent.EPS.in.the.Last.Four.Seasons -124552284 <2e-16 ***
## Cash.Flow.Per.Share 57403862 <2e-16 ***
## Revenue.Per.Share..Yuan.Â.. 75876487 <2e-16 ***
## Operating.Profit.Per.Share..Yuan.Â.. -124361620 <2e-16 ***
## Per.Share.Net.profit.before.tax..Yuan.Â.. 31107919 <2e-16 ***
## Realized.Sales.Gross.Profit.Growth.Rate 46076620 <2e-16 ***
## Operating.Profit.Growth.Rate -127836823 <2e-16 ***
## After.tax.Net.Profit.Growth.Rate -1183757 <2e-16 ***
## Regular.Net.Profit.Growth.Rate 9041319 <2e-16 ***
## Continuous.Net.Profit.Growth.Rate 17834207 <2e-16 ***
## Total.Asset.Growth.Rate 140962057 <2e-16 ***
## Net.Value.Growth.Rate 35763516 <2e-16 ***
## Total.Asset.Return.Growth.Rate.Ratio -23168127 <2e-16 ***
## Cash.Reinvestment.. 104007432 <2e-16 ***
## Current.Ratio -39486446 <2e-16 ***
## Quick.Ratio -156613086 <2e-16 ***
## Interest.Expense.Ratio -118220499 <2e-16 ***
## Total.debt.Total.net.worth 37948302 <2e-16 ***
## Debt.ratio.. 161136597 <2e-16 ***
## Net.worth.Assets NA NA
## Long.term.fund.suitability.ratio..A. 33327043 <2e-16 ***
## Borrowing.dependency 63864379 <2e-16 ***
## Contingent.liabilities.Net.worth -23691249 <2e-16 ***
## Operating.profit.Paid.in.capital 135987347 <2e-16 ***
## Net.profit.before.tax.Paid.in.capital -1557518 <2e-16 ***
## Inventory.and.accounts.receivable.Net.value -155216042 <2e-16 ***
## Total.Asset.Turnover 1340304 <2e-16 ***
## Accounts.Receivable.Turnover -242377587 <2e-16 ***
## Average.Collection.Days -145011977 <2e-16 ***
## Inventory.Turnover.Rate..times. -197757140 <2e-16 ***
## Fixed.Assets.Turnover.Frequency 148459224 <2e-16 ***
## Net.Worth.Turnover.Rate..times. -183083657 <2e-16 ***
## Revenue.per.person 100604097 <2e-16 ***
## Operating.profit.per.person -59653468 <2e-16 ***
## Allocation.rate.per.person -63067136 <2e-16 ***
## Working.Capital.to.Total.Assets -1213320 <2e-16 ***
## Quick.Assets.Total.Assets 24880846 <2e-16 ***
## Current.Assets.Total.Assets 1213319 <2e-16 ***
## Cash.Total.Assets -138381307 <2e-16 ***
## Quick.Assets.Current.Liability -54302467 <2e-16 ***
## Cash.Current.Liability 24042724 <2e-16 ***
## Current.Liability.to.Assets -1213320 <2e-16 ***
## Operating.Funds.to.Liability 109438597 <2e-16 ***
## Inventory.Working.Capital 29298193 <2e-16 ***
## Inventory.Current.Liability -29051800 <2e-16 ***
## Current.Liabilities.Liability 97451990 <2e-16 ***
## Working.Capital.Equity 130081184 <2e-16 ***
## Current.Liabilities.Equity 117042372 <2e-16 ***
## Long.term.Liability.to.Current.Assets -15747335 <2e-16 ***
## Retained.Earnings.to.Total.Assets 59207193 <2e-16 ***
## Total.income.Total.expense 783703 <2e-16 ***
## Total.expense.Assets 114384412 <2e-16 ***
## Current.Asset.Turnover.Rate 137086315 <2e-16 ***
## Quick.Asset.Turnover.Rate -58683491 <2e-16 ***
## Working.capitcal.Turnover.Rate -59440091 <2e-16 ***
## Cash.Turnover.Rate -147934153 <2e-16 ***
## Cash.Flow.to.Sales 3247095 <2e-16 ***
## Fixed.Assets.to.Assets 161895554 <2e-16 ***
## Current.Liability.to.Liability NA NA
## Current.Liability.to.Equity NA NA
## Equity.to.Long.term.Liability 52922374 <2e-16 ***
## Cash.Flow.to.Total.Assets 131900646 <2e-16 ***
## Cash.Flow.to.Liability -134672516 <2e-16 ***
## CFO.to.Assets -27507102 <2e-16 ***
## Cash.Flow.to.Equity -136952103 <2e-16 ***
## Net.Income.to.Total.Assets -351446409 <2e-16 ***
## No.credit.Interval -94577643 <2e-16 ***
## Gross.Profit.to.Sales -30117980 <2e-16 ***
## Net.Income.to.Stockholder.s.Equity 81891302 <2e-16 ***
## Liability.to.Equity -44734287 <2e-16 ***
## Degree.of.Financial.Leverage..DFL. -60472808 <2e-16 ***
## Interest.Coverage.Ratio..Interest.expense.to.EBIT. -6863935 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 13724 on 9899 degrees of freedom
## Residual deviance: 113465 on 9812 degrees of freedom
## AIC: 113641
##
## Number of Fisher Scoring iterations: 25Model Analysis
Model 1
logit_1.preds<-as.vector(ifelse(predict(logit_1,test_set)>0.5,"YES","NO"))## Warning in predict.lm(object, newdata, se.fit, scale = 1, type = if (type == :
## prediction from a rank-deficient fit may be misleadingcm1<-confusionMatrix(data=as.factor(logit_1.preds),test_set$Bankrupt)
cm1## Confusion Matrix and Statistics
##
## Reference
## Prediction NO YES
## NO 1593 37
## YES 56 18
##
## Accuracy : 0.9454
## 95% CI : (0.9336, 0.9557)
## No Information Rate : 0.9677
## P-Value [Acc > NIR] : 1.00000
##
## Kappa : 0.2513
##
## Mcnemar's Test P-Value : 0.06197
##
## Sensitivity : 0.9660
## Specificity : 0.3273
## Pos Pred Value : 0.9773
## Neg Pred Value : 0.2432
## Prevalence : 0.9677
## Detection Rate : 0.9349
## Detection Prevalence : 0.9566
## Balanced Accuracy : 0.6467
##
## 'Positive' Class : NO
## #print(c(paste0("Accuracy of LR with all features:",mean(logit_1.preds == test_set$Bankrupt))))
#print(c(paste0("F1-score of LR with all features:",F1_Score(logit_1.preds,test_set$Bankrupt))))
resultsCmp <- tibble(model = "Model 1",F1 = cm1$byClass[7],
deviance=logit_1$deviance,
Accuracy = cm1$overall[1],
Sensitivity = cm1$byClass[1],
Specificity = cm1$byClass[2],
Precision = cm1$byClass[5],
r2 = 1 - logit_1$deviance/logit_1$null.deviance,
AIC=logit_1$aic)As we can see the this model training with all features has accuracy 0.9454 and F-1 score of 0.971, although the accuracy looks high but when we focus on the confusion matrix we can find the specificity is only 0.3273 (means we are missing real bankrupting companies), our main purpose for this prediction is to help do Bankrupt company prediction. We should aim to increase the specificity so the model won’t miss real bankrupting companies.
Model 2
logit_3.preds<-as.vector(ifelse(predict(logit_3,test_set)>0.5,"YES","NO"))
cm2<-confusionMatrix(data=as.factor(logit_3.preds),test_set$Bankrupt)
cm2## Confusion Matrix and Statistics
##
## Reference
## Prediction NO YES
## NO 1635 47
## YES 14 8
##
## Accuracy : 0.9642
## 95% CI : (0.9543, 0.9725)
## No Information Rate : 0.9677
## P-Value [Acc > NIR] : 0.8148
##
## Kappa : 0.1929
##
## Mcnemar's Test P-Value : 4.182e-05
##
## Sensitivity : 0.9915
## Specificity : 0.1455
## Pos Pred Value : 0.9721
## Neg Pred Value : 0.3636
## Prevalence : 0.9677
## Detection Rate : 0.9595
## Detection Prevalence : 0.9871
## Balanced Accuracy : 0.5685
##
## 'Positive' Class : NO
## resultsCmp <- rbind(resultsCmp,tibble(model = "Model 2",
F1 = cm2$byClass[7],
deviance = logit_3$deviance,
Accuracy = cm2$overall[1],
Sensitivity = cm2$byClass[1],
Specificity = cm2$byClass[2],
Precision = cm2$byClass[5],
r2 = 1 - logit_3$deviance / logit_3$null.deviance,
AIC = logit_3$aic))We applied backward stepwise regression on the Model 1 to build this Model 2. Even though the accuracy has improved and AIC has reduced, but the specificity has drastically gone down (14.54%) which is not good.
Model 3
logit_1_up.preds<-as.vector(ifelse(predict(logit_1_up,test_set)>0.5,"YES","NO"))## Warning in predict.lm(object, newdata, se.fit, scale = 1, type = if (type == :
## prediction from a rank-deficient fit may be misleadingcm3<-confusionMatrix(data=as.factor(logit_1_up.preds),test_set$Bankrupt)
cm3## Confusion Matrix and Statistics
##
## Reference
## Prediction NO YES
## NO 1497 15
## YES 152 40
##
## Accuracy : 0.902
## 95% CI : (0.8869, 0.9157)
## No Information Rate : 0.9677
## P-Value [Acc > NIR] : 1
##
## Kappa : 0.2882
##
## Mcnemar's Test P-Value : <2e-16
##
## Sensitivity : 0.9078
## Specificity : 0.7273
## Pos Pred Value : 0.9901
## Neg Pred Value : 0.2083
## Prevalence : 0.9677
## Detection Rate : 0.8785
## Detection Prevalence : 0.8873
## Balanced Accuracy : 0.8175
##
## 'Positive' Class : NO
## resultsCmp <- rbind(resultsCmp,tibble(model = "Model 3",
F1 = cm3$byClass[7],
deviance = logit_1_up$deviance,
Accuracy = cm3$overall[1],
Sensitivity = cm3$byClass[1],
Specificity = cm3$byClass[2],
Precision = cm3$byClass[5],
r2 = 1 - logit_1_up$deviance / logit_1_up$null.deviance,
AIC = logit_1_up$aic))Considering the imbalanced nature of test data, we did up sampling of the training data set and generated the model 3.
Model 4
logit_3_up.preds<-as.vector(ifelse(predict(logit_3_up,test_set)>0.5,"YES","NO"))## Warning in predict.lm(object, newdata, se.fit, scale = 1, type = if (type == :
## prediction from a rank-deficient fit may be misleadingcm4<-confusionMatrix(data=as.factor(logit_3_up.preds),test_set$Bankrupt)
cm4## Confusion Matrix and Statistics
##
## Reference
## Prediction NO YES
## NO 1473 17
## YES 176 38
##
## Accuracy : 0.8867
## 95% CI : (0.8707, 0.9014)
## No Information Rate : 0.9677
## P-Value [Acc > NIR] : 1
##
## Kappa : 0.2437
##
## Mcnemar's Test P-Value : <2e-16
##
## Sensitivity : 0.8933
## Specificity : 0.6909
## Pos Pred Value : 0.9886
## Neg Pred Value : 0.1776
## Prevalence : 0.9677
## Detection Rate : 0.8644
## Detection Prevalence : 0.8744
## Balanced Accuracy : 0.7921
##
## 'Positive' Class : NO
## resultsCmp <- rbind(resultsCmp,tibble(model = "Model 4",
F1 = cm4$byClass[7],
deviance = logit_3_up$deviance,
Accuracy = cm4$overall[1],
Sensitivity = cm4$byClass[1],
Specificity = cm4$byClass[2],
Precision = cm4$byClass[5],
r2 = 1 - logit_3_up$deviance / logit_3_up$null.deviance,
AIC = logit_3_up$aic))We applied backward stepwise regression model 3 to built model 4.
Model Comparison
#resultsCmp
kbl(resultsCmp) %>%
kable_styling(bootstrap_options = c("striped", "hover", "condensed"))| model | F1 | deviance | Accuracy | Sensitivity | Specificity | Precision | r2 | AIC |
|---|---|---|---|---|---|---|---|---|
| Model 1 | 0.9716377 | 17877.6521 | 0.9454225 | 0.9660400 | 0.3272727 | 0.9773006 | -11.2631517 | 18059.6521 |
| Model 2 | 0.9816872 | 821.2444 | 0.9642019 | 0.9915100 | 0.1454545 | 0.9720571 | 0.4366685 | 879.2444 |
| Model 3 | 0.9471686 | 116709.3497 | 0.9019953 | 0.9078229 | 0.7272727 | 0.9900794 | -7.5038384 | 116891.3497 |
| Model 4 | 0.9385155 | 113465.4209 | 0.8867371 | 0.8932686 | 0.6909091 | 0.9885906 | -7.2674747 | 113641.4209 |
Model Selection
We could see that Model1 and Model2 have high F1,Accuracy,Sensitivity but the specificity values are really low(32 % and 14.54%). Hence we could reject both the Model 1 and Model 2.
Our earlier analysis showed that the data set has only 3% cases where bankruptcy has happened. Due to the imbalance nature of the data set , we did up sampling of the data and developed the model 3. We could see that Specificity has improved to 72.27%.
Model 4 was built by applying stepwise regression on Model 3. Even though model 4 has lower AIC, model 3 has better precision and sensitivity and accuracy. Hence model 3 is our choice.