German Credit Analysis from https://archive.ics.uci.edu/ml/datasets/Statlog+%28German+Credit+Data%29

The traget of the analysis is to classify customers as good or bad based on attributes given ionm the dataset by creating a classification model.

Description of the German credit dataset.

1. Title: German Credit data

2. Source Information

Professor Dr. Hans Hofmann
Institut f"ur Statistik und "Okonometrie
Universit"at Hamburg
FB Wirtschaftswissenschaften
Von-Melle-Park 5
2000 Hamburg 13

3. Number of Instances: 1000

Two datasets are provided. the original dataset, in the form provided by Prof. Hofmann, contains categorical/symbolic attributes and is in the file "german.data".

For algorithms that need numerical attributes, Strathclyde University produced the file "german.data-numeric". This file has been edited and several indicator variables added to make it suitable for algorithms which cannot cope with categorical variables. Several attributes that are ordered categorical (such as attribute 17) have been coded as integer. This was the form used by StatLog.

6. Number of Attributes german: 20 (7 numerical, 13 categorical) Number of Attributes german.numer: 24 (24 numerical)

7. Attribute description for german

Attribute 1: (qualitative) Status of existing checking account A11 : ... < 0 DM A12 : 0 <= ... < 200 DM A13 : ... >= 200 DM / salary assignments for at least 1 year A14 : no checking account

Attribute 2: (numerical) Duration in month

Attribute 3: (qualitative) Credit history A30 : no credits taken/ all credits paid back duly A31 : all credits at this bank paid back duly A32 : existing credits paid back duly till now A33 : delay in paying off in the past A34 : critical account/ other credits existing (not at this bank)

Attribute 4: (qualitative) Purpose A40 : car (new) A41 : car (used) A42 : furniture/equipment A43 : radio/television A44 : domestic appliances A45 : repairs A46 : education A47 : (vacation - does not exist?) A48 : retraining A49 : business A410 : others

Attribute 5: (numerical) Credit amount

Attibute 6: (qualitative) Savings account/bonds A61 : ... < 100 DM A62 : 100 <= ... < 500 DM A63 : 500 <= ... < 1000 DM A64 : .. >= 1000 DM A65 : unknown/ no savings account

Attribute 7: (qualitative) Present employment since A71 : unemployed A72 : ... < 1 year A73 : 1 <= ... < 4 years
A74 : 4 <= ... < 7 years A75 : .. >= 7 years

Attribute 8: (numerical) Installment rate in percentage of disposable income

Attribute 9: (qualitative) Personal status and sex A91 : male : divorced/separated A92 : female : divorced/separated/married A93 : male : single A94 : male : married/widowed A95 : female : single

Attribute 10: (qualitative) Other debtors / guarantors A101 : none A102 : co-applicant A103 : guarantor

Attribute 11: (numerical) Present residence since

Attribute 12: (qualitative) Property A121 : real estate A122 : if not A121 : building society savings agreement/ life insurance A123 : if not A121/A122 : car or other, not in attribute 6 A124 : unknown / no property

Attribute 13: (numerical) Age in years

Attribute 14: (qualitative) Other installment plans A141 : bank A142 : stores A143 : none

Attribute 15: (qualitative) Housing A151 : rent A152 : own A153 : for free

Attribute 16: (numerical) Number of existing credits at this bank

Attribute 17: (qualitative) Job A171 : unemployed/ unskilled - non-resident A172 : unskilled - resident A173 : skilled employee / official A174 : management/ self-employed/ highly qualified employee/ officer

Attribute 18: (numerical) Number of people being liable to provide maintenance for

Attribute 19: (qualitative) Telephone A191 : none A192 : yes, registered under the customers name

Attribute 20: (qualitative) foreign worker A201 : yes A202 : no

8. Cost Matrix

This dataset requires use of a cost matrix with

the rows represent the actual classification and the columns the predicted classification.

It is worse to class a customer as good when they are bad (5), than it is to class a customer as bad when they are good (1).

Cleaning the data

I am using credit.data file for the analysis. After loading the file into environment the column names are changed to more appropriate names for better understanding and easier analysis. After that cconverted (1,2) to (0,1) in the columns for easier analysis. Then check for NA values, as there is none, no imputaion is reuired.

Plotting univariate graphs

Plotting variables to gain overall idea about them. Also, cleaning numerical variable with outliers.

Plotting some bivarite analysis

Doing some bivariate analysis and plotting them to gain idea about interdependence about hose variables.

Model creation

I decided to create model using general linear model. For general linear model as variables are not in much diverse range except 2/3 variable so I decided scaling is not required. However this may have effected the model so it needs to be addressed in future evaluation.

Cost matrix

As per the instruction in the given dataset I am using the following cost matrix to estimate the cost of prediction. 1 2

1 0 1

2 5 0

(1 = Good, 2 = Bad)

General linear model

For general linear model I kept all the variables in numerical format and split the data and 70% and 30% ratio for training and testing purpose respectively. Step-1 Create model with all other features vs classification and look at the summary. Step-2 Step wise reduce AIC using stepAIC function from MASS package and store it in a variable. Step-3 Call the formula predicted by stepAIC to bulid next model. View the summary and vif of the model. Step-4 Remove variable with highest vif and not much significant p-value and create a model with the new variables. At each point remove only one variable and check summary and vif. After all the variable are having less than 2 vif stop checking for vif anymore. Step-5 After that start looking at varibale with high p-value and remove one at a time and check summary. Contin ue to do so until all the variables have low p-value and/or number of variables are low. Step-6 Build the final model based on the variable with low p-values and low vif. Step-7 After that predict classification values of test data set. Use a cutoff of 0.5 to predict as the model gives only probability values. So, values greater than equal to 0.5 should be predicted as 1 i.e. prone to default and values lower than 0.5 shoulds be predicted as 0 i.e. not prone to default. cost calculation After that create a confusion matridx showing values of predicted and actual values with false-positive + true-positive + true-negative + false-negative. Then Mutliply that confusion matrix with the cost matrix appropriately to get the cost of the model i.e. what is the cost for predicting using the model.

Predicting all the values

After the final model creation just get all the predictions and append them in a column.