32 References [1] E. D. Madyatmadja and M. Aryuni, “Comparative Study of Data Mining Model for Credit Card Application Scoring in Bank,” J. Theor. Appl. Inf. Technol., vol. 59, no. 2, 2014. [2] M. Aryuni and E. D. Madyatmadja, “Feature Selection in Credit Scoring Model for Credit Card Applicants in XYZ Bank: A Comparative Study,” Int. J. Multimed. Ubiquitous Eng., vol. 10, no. 5, pp. 17–24, May 2015. [3] M. Nithya, S. Suba, and B. Vaishnavi, “Predict the usage of laptops among students in rural areas using weka tool,” Int. J. Adv. Technol. Eng. Sci., vol. Vol. No.3, no. 01 September 2015, p. 6. [4] N. Padhy, “Data Mining: A prediction Technique for the workers in the PR Department of Orissa (Block and Panchayat),” Int. J. Comput. Sci. Eng. Inf. Technol., vol. 2, no. 5, pp. 197–36, Oct. 2012. [5] D. Kumar Thiwary, “A Comparative Study Of Classification Algorithms For Credit Card Approval Using WEKA,” GALAXY Int. Interdiscip. Res. J., vol. Vol.2, Mar. 2014. [6] Plc.lk, 'People's Leasing Company PLC - Annual Report 2013 - 2014', 2015. [Online]. Available: http://www.plc.lk/inpages/annual_report_2013_14/risk- management.html. [Accessed: 03- Jan- 2015]. [7]Annual Report 2013/14, 1st ed. colombo: Swarnamahal Financial Services PLC, 2015, pp. 23,24. [8]Annual Report 2012/13, 1st ed. colombo: Swarnamahal Financial Services PLC, 2014, p. 25 33 [9]S. Faizan Ahmed and Q. Ali Malik, 574 International Journal of Economics and Financial Issues | Vol 5 • Issue 2 • 2015 Credit Risk Management and Loan Performance: Empirical Investigation of Micro Finance Banks of Pakistan, 1st ed. International Journal of Economics and Financial Issues, 2015. [10]D. Foust and A. Pressman, 'Not-So-Magic Numbers', Business Week, 2008. [11]Foundations of Banking Risk: An Overview of Banking, Banking Risks, and Risk- Based Banking Regulation., 1st ed. New Jersey: John Wiley& Sons, 2009. [12]T. Gestel and B. Baesens, Credit Risk Management: Basic Concepts: Financial Risk Components, rating analysis,models,economic and regulary capilal, 1st ed. Oxford University Press, 2008. [13]G. Yingjian and W. Chong, Research on Credit Risk Assessment in Commercial Bank Based on Information Integration, 1st ed. . 34 Appendix A: Categorization Data sets have been categorized on to following order •Dependent Variable: Status - Ceased, Closed •Other Variables: Factor 1 - Profession - Businessmen, Doctor, Engineer, Executive, Farmer, Lawyer, Lecturer, Manager, Other, Self Employee, Small Businessmen, Teacher Factor 2 - Brand Name - Bajaj, Chevrolet, Isuzu, Mahindra & Mahindra, Maruti Suzuki, Mazda, Mitsubishi Motors, Nissan, Suzuki, Tafe, Tata Motors, Toyota Factor 3 - Model Name - 800, 45D, ACE, ALTO, AR4S-UG, ATLAS, AUTO 4S, AUTO AR4S, AUTO RE 2 STROKE, BAJAJ 4 S, BAJAJ AUTO, BOLERO MAXI TRUCK, CONDOR, COROLLA, DBA-NZE141, DYNA, ELF 350, HIACE, KF- GM70-HALF-BODY, KG-LH172, KG-VWE25, LA-HA238, LP7155, PAJERO JEEP, RE 205, TITAN, TOWNACE, UA-HR528, VANATTE Factor 4 - Manufacture Year - 1993 to 2012 Factor 5 - Vehicle Class - Dual purpose Motor vehicle, Farm vehicle, Heavy Motor Lorry, Light Motor Lorry, Motor Car, Motor Coach, Motor Lorry UP 1700kg, Motor Tricycle, Motorcycles UP 100CC Factor 6 - Fuel Type - Diesel, Petrol Factor 7 - Actual Value - <5,00,000(P), 5,00,000-9,99,999(Q), 10,00,000- 14,99,999(R), 15,00,000-19,99,999(S), 20,00,000-24,99,999(T), 25,00,000- 29,99,999(U), 30,00,000-34,99,999(V), >35,00,000(W) Factor 8 - Lease Percentage - 60% to 100% Factor 9 - Req. Amount - <5,00,000(p), 5,00,000-9,99,999(q), 10,00,000-14,99,999(r), 15,00,000-19,99,999(s), 20,00,000-24,99,999(t), 25,00,000-29,99,999(u), 30,00,000- 34,99,999(v), >35,00,000(w) 35 Factor 10 - Monthly Income - <50,000(I), 50,000-99,999(II), 1,00,000-1,49,999(III), 1,50,000-1,99,999(IV), >2,00,000(V) Factor 11 - Interest - 9 to 15 Factor 12 - Number of rentals - 12 to 60 Factor 13 - Installment - <20,000(i), 20,000-39,999(ii), 40,000-59,999(iii), 60,000- 79,999(iv), 80,000-99,999(v), >1,00,000(vi) 36 Appendix B: Variable Selection Procedure (Technology: Chi – Squared) Status * Gender Chi-Square Tests Value df Asymp. Sig. (2- sided) Exact Sig. (2- sided) Exact Sig. (1- sided) Pearson Chi-Square 2.691a 1 .101 Continuity Correctionb 2.572 1 .109 Likelihood Ratio 2.700 1 .100 Fisher's Exact Test .107 .054 N of Valid Cases 6000 a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 393.38. b. Computed only for a 2x2 table Status * Age Chi-Square Tests Value df Asymp. Sig. (2- sided) Pearson Chi-Square 3.878a 6 .693 Likelihood Ratio 3.877 6 .693 N of Valid Cases 6000 a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 16.91. 37 Status * Profession Chi-Square Tests Value df Asymp. Sig. (2- sided) Pearson Chi-Square 679.354a 11 .000 Likelihood Ratio 955.310 11 .000 N of Valid Cases 6000 a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 29.82. Status * District Chi-Square Tests Value df Asymp. Sig. (2- sided) Pearson Chi-Square 4.115a 7 .766 Likelihood Ratio 4.109 7 .767 N of Valid Cases 6000 a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 44.06. 38 Status * BrandName Chi-Square Tests Value df Asymp. Sig. (2- sided) Pearson Chi-Square 157.472a 11 .000 Likelihood Ratio 167.321 11 .000 N of Valid Cases 6000 a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 13.35. Status * ModelName Chi-Square Tests Value df Asymp. Sig. (2- sided) Pearson Chi-Square 392.574a 28 .000 Likelihood Ratio 457.831 28 .000 N of Valid Cases 6000 a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 12.02. 39 Status * Manuf.Year Chi-Square Tests Value df Asymp. Sig. (2- sided) Pearson Chi-Square 41.545a 19 .002 Likelihood Ratio 41.803 19 .002 N of Valid Cases 6000 a. 1 cells (2.5%) have expected count less than 5. The minimum expected count is 4.90. Status * Vehicle Class Chi-Square Tests Value df Asymp. Sig. (2- sided) Pearson Chi-Square 252.964a 8 .000 Likelihood Ratio 269.882 8 .000 N of Valid Cases 6000 a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 13.35. 40 Status * Fuel Type Chi-Square Tests Value df Asymp. Sig. (2- sided) Exact Sig. (2- sided) Exact Sig. (1- sided) Pearson Chi-Square 37.032a 1 .000 Continuity Correctionb 36.705 1 .000 Likelihood Ratio 36.967 1 .000 Fisher's Exact Test .000 .000 N of Valid Cases 6000 a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 987.90. b. Computed only for a 2x2 table Status * ActualValue Chi-Square Tests Value df Asymp. Sig. (2- sided) Pearson Chi-Square 311.275a 7 .000 Likelihood Ratio 367.826 7 .000 N of Valid Cases 6000 a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 99.68. 41 Status * Lease Percentage Chi-Square Tests Value df Asymp. Sig. (2- sided) Pearson Chi-Square 27.255a 4 .000 Likelihood Ratio 27.262 4 .000 N of Valid Cases 6000 a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 494.40. Status * Req.Amount Chi-Square Tests Value df Asymp. Sig. (2- sided) Pearson Chi-Square 197.350a 7 .000 Likelihood Ratio 208.573 7 .000 N of Valid Cases 6000 a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 86.33. 42 Status * Monthly Income Chi-Square Tests Value df Asymp. Sig. (2- sided) Exact Sig. (2- sided) Exact Sig. (1- sided) Pearson Chi-Square 3.921a 1 .048 Continuity Correctionb 3.571 1 .059 Likelihood Ratio 3.892 1 .049 Fisher's Exact Test .057 .030 N of Valid Cases 6000 a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 56.07. b. Computed only for a 2x2 table Status * Interest Chi-Square Tests Value df Asymp. Sig. (2- sided) Pearson Chi-Square 134.384a 2 .000 Likelihood Ratio 138.243 2 .000 N of Valid Cases 6000 a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 561.59. 43 Status * No.of Rentals Chi-Square Tests Value df Asymp. Sig. (2- sided) Pearson Chi-Square 88.824a 4 .000 Likelihood Ratio 90.264 4 .000 N of Valid Cases 6000 a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 425.42. Status * Installment Chi-Square Tests Value df Asymp. Sig. (2-sided) Pearson Chi-Square 85.487a 4 .000 Likelihood Ratio 87.334 4 .000 N of Valid Cases 6000 44 Appendix C: Best Model Selection ADTree Correctly Classified Instances Time Taken to Build the Model 58.3% 0.22 Seconds 45 BFTree Correctly Classified Instances Time Taken to Build the Model 59.25% 31.04 Seconds 46 FT Tree Correctly Classified Instances Time Taken to Build the Model 75.65 % 12.61 Seconds 47 J48 Correctly Classified Instances Time Taken to Build the Model 78.4% 86.52 Seconds 48 J48graft Correctly Classified Instances Time Taken to Build the Model 78.4% 98.1 Seconds 49 LAD Tree Correctly Classified Instances Time Taken to Build the Model 58.1% 5.35 Seconds 50 LMT Correctly Classified Instances Time Taken to Build the Model 58.6% 89.46 Seconds 51 RandomTree Correctly Classified Instances Time Taken to Build the Model 83.65% 0.07 Seconds 52 SimpleCart Correctly Classified Instances Time Taken to Build the Model 60.2 % 12.02 Seconds 53 Correctly Classified Instances Time Taken to Build the Model 83.65 % 5.42 Seconds