Doctor of philosophy

Table - 29

Amount Overdue

		Frequency	Percent	Valid Percent	Cumulative Percent
Valid	Rs. 0.00	6508	54.0	54.0	54.0
	Rs. 1 to rs. 4000	820	6.8	6.8	60.8
	Rs. 4001 to Rs. 6000	1634	13.6	13.6	74.4
	Rs. 6001 to Rs. 8000	1076	8.9	8.9	83.3
	Rs. 8001 to Rs. 10000	670	5.6	5.6	88.8
	Above Rs. 10000	1345	11.2	11.2	100.0
	Total	12053	100.0	100.0

Amout overdue varies from one EMI to more than 10 EMIs.

Table - 30

No. of Overdues

		Frequency	Percent	Valid Percent	Cumulative Percent
Valid	0	6508	54.0	57.4	57.4
	1 to 3	1675	13.9	14.8	72.1
	4 to 6	2050	17.0	18.1	90.2
	7 to 9	645	5.4	5.7	95.9
	10 to 12	256	2.1	2.3	98.2
	Above 12	209	1.7	1.8	100.0
	Total	11343	94.1	100.0
Missing	System	710	5.9
Total		12053	100.0

Table - 31

Customer Classification

		Frequency	Percent	Valid Percent	Cumulative Percent
Valid	Bad Customer	5545	46.0	46.0	46.0
	Good Customer	6508	54.0	54.0	100.0
	Total	12053	100.0	100.0

There are more good customers than bad customers but there is only a marginal difference.

6.1 Relationship of background Variables With Customer Type (Good or Bad):

Table - 32

Age Vs Customer Type





Fig-31 Age-Customer classification

From the table and the chart above, it can be seen that the customers in the age group of 26 to 35 are more likely to default than the customers in the age group of 46 and above. The chi-squared value of 605 at 4 degrees of freedom is statistically significant at 99.9 confidence level, implying that age and customer type are related.

Table - 33

Educational Qualification Vs Customer Type





Fig-32 Qualification-Customer classification

The less educated and those with PG qualification are less likely to default than those with school and degree qualification. The chi-squared value of 101 at 4 degrees of freedom is statistically significant at 99.99% confidence level, implying that educational qualification and customer type are related.

Table - 34

Number of dependents Vs Customer Type





Fig-33 No of dependents-Customer classification

Customers with 2 or less dependents are more likely to default that those with 3 or more dependents. The chi-squared value of 269 at 6 degrees of freedom is statistically significant at 99.99% confidence level, implying that number of dependents and customer type are related.

Table - 35

Income Vs Customer Type





Fig-34 Income –Customer classsification

Customers in the Rs. 7000 to 9000 income group are more likely to default than those in other income groups. The chi-squared value of 74 at 5 degrees of freedom is statistically significant at 99.99% confidence level, implying that income and customer type are related.

Table - 36

Other Income Vs Customer Type





Fig-35 Other income- Customer classification

As other income increases the likelyhood of default decreases. The chi-squared value of 71 at 3 degrees of freedom is statistically significant at 99.99% confidence level, implying that other income and customer type are related.

Table - 37

Years of Experience Vs Customer Type





Fig-36 Experience – Customer classification

Customers in the 5 to 8 years experience group are more likely to default than others. The chi-squared value of 69 at 6 degrees of freedom is statistically significant at 99.99% confidence level, implying that other income and customer type are related.

Table - 38

Type of house Vs Customer Type





Fig-37 Type of house-Customer classification

Customers living in their own house are more likely to default than those in rented or leased houses. The chi-squared value of 23 at 1 degrees of freedom is statistically significant at 99.99% confidence level, implying that other income and customer type are related.

Table - 39

Rent Paid Vs Customer Type





Fig-38 Rent-Customer Classification

There is no significant relationship that could be established between rent amount paid and the customer type. The chi-squared value of 9.8 at 5 degrees of freedom is not statistically significant at (p value = 0.08), implying that rent paid and customer type are related.

Table - 40

Down Payment * Customer Classification Crosstabulation

			Customer Classification		Total
			Bad Customer	Good Customer	Bad Customer
Down Payment	Less than Rs. 5000	Count	956	341	1297
		% within Down Payment	73.7%	26.3%	100.0%
	Rs. 5001 to Rs. 8000	Count	1567	1510	3077
		% within Down Payment	50.9%	49.1%	100.0%
	Rs. 8001 to Rs 11000	Count	1078	1711	2789
		% within Down Payment	38.7%	61.3%	100.0%
	Rs. 11001 to Rs. 14000	Count	751	945	1696
		% within Down Payment	44.3%	55.7%	100.0%
	Rs. 14001 to Rs. 17000	Count	541	563	1104
		% within Down Payment	49.0%	51.0%	100.0%
	Rs. 17001 to Rs. 20000	Count	392	390	782
		% within Down Payment	50.1%	49.9%	100.0%
	Above Rs. 20000	Count	259	1047	1306
		% within Down Payment	19.8%	80.2%	100.0%
Total		Count	5544	6507	12051
		% within Down Payment	46.0%	54.0%	100.0%

Chi-Square Tests

	Value	df	Asymp. Sig. (2-sided)
Pearson Chi-Square	862.985(a)	6	.000
Likelihood Ratio	904.867	6	.000
Linear-by-Linear Association	450.784	1	.000
N of Valid Cases	12051

0 cells (.0%) have expected count less than 5. The minimum expected count is 359.76.



Fig-39 Downpayment-Customer classification

Down payment has significant relationship with the class of customers as reflected in the Chi-square value of 862.985 at 6 degrees of freedom. Customers who have paid Rs 20000 and more have a very good Repayment behavior.

Table - 41

TV Ownership * Customer Classification Crosstabulation

			Customer Classification		Total
			Bad Customer	Good Customer	Bad Customer
TV Ownership	No	Count	1103	830	1933
		% within TV Ownership	57.1%	42.9%	100.0%
	Yes	Count	4442	5678	10120
		% within TV Ownership	43.9%	56.1%	100.0%
Total		Count	5545	6508	12053
		% within TV Ownership	46.0%	54.0%	100.0%