Credit Scoring and Pattern in the US - Research Paper Example

Credit scoring and pattern in the US US economy depends on loans to grow individuals and firms, in terms of the financial health of the individual or the firm subject to the risks evaluated by the funder or the loan investor. This credit reference data are freely available online but the individual names and firms are deleted for privacy purposes. Introduction We conducted of loans evaluations in the US, the data was obtained from the credit reference bureau and covered a period of 2007 – 2011. The loan funders must evaluate the risks involved and the sum up the probability of uncertain occurrences that affect the loan repayment and defaulting (Pritchard, 2015). Literature review The financial institutions and the banks must evaluate the creditworthy of the applicants and distinguish them from the defaulters, thus each loan applicant must evaluated on capital or income, capacity to pay the loan, past and present character and collaterals (A. I. Marqués, 2012). Thus in the recent history the financial institutions are involved in credit assessment and risk prediction. Credit scoring come along the strong benefits that the financial institution rely upon in making the decision, hence this model requires less input of information in variable form which are statistically evaluated on their significance and correlation to determine the repayment performance (Hussein A. Abdou, 2011). Credit scoring is a model that links the variables in financial sector to determine the loan default probability, thus is a predictor risk that comes forth in determining when the individuals and firms are on the verge of bankruptcy (Asia Samreen, 2013). Research question Do loan approvals vary with states and the credit rating of the individual determines the loan amount. Hypothesis Ho = The credit scoring of the individual determines the loan amount to be secured. Ha = The credit scoring of an individual does not determine the loan amount to be secured. Methodology We obtained secondary data from Lending Club Statistics, the data time series secondary data. The data has a population of 42540 who all had an equal chance to be sampled for our analysis (Lending Club, 2015). Data Analysis We conducted an evaluation of the Lending Club data in determining the credits risks, approvals and the loan evaluation. We subjected the data to various measures of spread, the data was first subjected to random selection was conducted on the data, and a sample of N= 50 was selected for analysis Descriptive statistics The graph indicates that 48% of US loan seekers have mortgages there houses, with 10% own house they occupy. The graph indicates that 76% of the loan applicants qualify for loans repayable in 3 years and below. The graph indicates that 18% of loans applicants are in the state of Califonia and 2% of the applicants are from Arkansas. Table 1 Descriptive Statistics The total amount committed to that loan at that point in time. The annual income provided by the borrower during registration. Interest received to date N Valid 50 50 50 Missing 0 0 0 Mean 11260.50 71554.18 2343.0202 Std. Error of Mean 1109.792 7576.078 376.13325 Median 12000.00 59000.00 1267.8400 Mode 12000 60000 92.73a Std. Deviation 7847.416 53570.958 2659.66374 Variance 61581941.071 2869847556.967 7073811.200 Skewness .905 1.960 1.497 Std. Error of Skewness .337 .337 .337 Kurtosis .768 3.936 1.286 Std. Error of Kurtosis .662 .662 .662 Range 33400 241588 10325.18 Minimum 1600 8412 92.73 Maximum 35000 250000 10417.91 Percentiles 25 4150.00 39375.00 426.7425 50 12000.00 59000.00 1267.8400 75 16000.00 80500.00 3047.5325 a. Multiple modes exist. The smallest value is shown The descriptive statistics of annual income, interest paid and loan given to the client are computed in Table 1.The maximum annual income is $250000 and the minimum income is $8412. The data is Skewness and Kurtosis indicate > 1 thus the income data is skewed to the right the normal distribution. The average income of the loans applicant is $71554, with mode of $60000 and median of $59000. The table indicates that average loan approved is $11260 attracting an average interest of $2343. The graph shows the total amount of loans commited and indicates show that the data is skewed to the right and is normally distributed. The graph indicates income has a normal distribution with leptokurtosis and skewed to the right. Table 2 Descriptive Statistics The total amount committed to that loan at that point in time. The annual income provided by the borrower during registration. Interest Rate on the loan Interest received to date Mean Mean Mean Mean The state provided by the borrower in the loan application Alabama 13708 45668 12.5733% 4052.68 Arkansas 29175 55000 10.6500% 7563.12 Arizona 14000 45000 13.0600% 2086.07 California 8711 72851 12.0011% 2210.71 Delaware 10625 39000 9.9900% 1131.81 Florida 5510 35720 12.9620% 1213.62 Georgia 5263 48200 10.8750% 1013.46 Kansas 6300 51500 5.7900% 574.88 Maine 3542 56167 6.5800% 353.06 Minnesota 6000 50400 6.9200% 380.87 New Jersey 11000 132000 12.2800% 1810.93 Nevada 14000 44000 20.3000% 6840.86 New York 17000 65758 11.5600% 2112.76 Ohio 13500 116000 9.7000% 2028.78 Pennsylvania 16000 150000 16.7000% 4442.98 South Carolina 5000 39500 7.7400% 437.05 Texas 9333 49000 10.2967% 1384.60 Virginia 22250 140155 12.5350% 5462.43 Washington 12240 88720 10.5000% 1671.55 Wisconsin 16250 44206 9.7800% 3109.24 Table 2 indicates that Arkansas received the highest amount of the loan amount at mean $29175 and as illustrated in the graph below.Arkansas loan applicants compromised 2% an indication that most the credit rating of the state is high. With Califonia having the highest loan applicants at 18% was among the state that received lowest loan amount, indicating that their are high rate of loan defaulters in Califonia thus credit scoring is low. The graph show the mount of income generated in states and the loan amount per states. We further subjected our tests, to evaluate the correlation between the loan repayments and the amount of the loan, revolving credit balance, revolving line of utilization and interest rate on loan. Table 3. Correlations Payments received to date for total amount funded Total credit revolving balance Revolving line utilization rate Interest Rate on the loan Payments received to date for total amount funded Pearson Correlation 1 .242 .298* .345* Sig. (2-tailed) .091 .036 .014 N 50 50 50 50 Total credit revolving balance Pearson Correlation .242 1 .475** .230 Sig. (2-tailed) .091 .000 .108 N 50 50 50 50 Revolving line utilization rate Pearson Correlation .298* .475** 1 .500** Sig. (2-tailed) .036 .000 .000 N 50 50 50 50 Interest Rate on the loan Pearson Correlation .345* .230 .500** 1 Sig. (2-tailed) .014 .108 .000 N 50 50 50 50 *. Correlation is significant at the 0.05 level (2-tailed). **. Correlation is significant at the 0.01 level (2-tailed). Correlation coefficients are as shown in Table 3 indicates that all the variables are statistically significant as p > .05 with all the variable showing a positive correlation with the payments. As shown in the three graphs below. Thus we can conclude that payments received as a great strong relationship with interest rate, credit revolving balance and revolving line. Table 4 Model Summary Model R R Square Adjusted R Square Std. Error of the Estimate 1 .501a .251 .147 8045.20434317225800 a. Predictors: (Constant), The total number of credit lines currently in the borrowers credit file, Interest Rate on the loan, A ratio calculated using the borrower’s total monthly debt payments , Total credit revolving balance, Revolving line utilization rate, The annual income provided by the borrower during registration. From the Table 4 R = .5 and R2 = 25.1%, indicating 25% variance payment received Table 5 ANOVAa Model Sum of Squares Df Mean Square F Sig. 1 Regression 934776921.102 6 155796153.517 2.407 .043b Residual 2783188455.706 43 64725312.923 Total 3717965376.809 49 a. Dependent Variable: Payments received to date for total amount funded b. Predictors: (Constant), The total number of credit lines currently in the borrowers credit file, Interest Rate on the loan, A ratio calculated using the borrower’s total monthly debt payments , Total credit revolving balance, Revolving line utilization rate, The annual income provided by the borrower during registration. The ANOVA results indicate that they are statistically significance,F(6,49) = 2.4, p = .04 with a low n2 = 25% as figured out in Table 5. Table 6 Coefficientsa Model Unstandardized Coefficients Standardized Coefficients t Sig. B Std. Error Beta 1 (Constant) -2815.688 4836.533 -.582 .563 Interest Rate on the loan 598.733 345.444 .265 1.733 .090 The annual income provided by the borrower during registration. .038 .028 .236 1.348 .185 A ratio calculated using the borrower’s total monthly debt payments 65.633 204.529 .049 .321 .750 Total credit revolving balance -.007 .102 -.012 -.071 .944 Revolving line utilization rate 32.024 51.117 .107 .626 .534 The total number of credit lines currently in the borrowers credit file 136.785 124.480 .170 1.099 .278 a. Dependent Variable: Payments received to date for total amount funded Multiple regression was carried out to evaluate the measures of strength predicted on payment received by financial institution, as indicated above R2 = 25.1% indicating a variance on payment received and the strength on the linear combination. From Table 6 we can get our regression equation model as follows Payment received = -2816 + 598 Interest rate + .038 Annual income + 66 Ratio calculated - .007 Total credit revolving bal + 32 revolving line +137 total number of credit lines. Hence we can draw a conclusion that the credit score of an individual depends on the credit lines is servicing in relation to the interest accrued in loans taken in the credit lines and th ratio calculated. If the above factor increases then the payments received increases putting the credit score in line. If the annual income decreases then the loan cannot be sufficed, lowering the credit score of the applicant. Conclusion Arkansas State has high loan applicants with high credit scoring, and if the payments received by the financiers is higher than the income source then the risk of having low credit score, hence higher risks of lending occurs. Most loans applicants prefer the loans of 3 years and below and most of them have mortgage there properties. Further the a ratio calculated using the past history record and thus the patterns that the financial sector use in rejecting the loan applications include the ratio calculated, the interest rate being paid and total number of credit lines. Bibliography A. I. Marqués, V. G. (2012). Evolutionary computing to credit scoring. Journal of the Operational Research Society, 1384–1399. Imechukuliwa toka Palgrave Macmillan: http://www.palgrave-journals.com/jors/journal/v64/n9/full/jors2012145a.html Asia Samreen, F. B. (2013). Forecasting Creditworthiness of Corporate Borrowers. International Journal of Business and Commerce, 1 - 26. Hussein A. Abdou, J. P. (2011). Credit Scoring, Statistical Techniques and Evaluation. Intelligent Systems in Accounting, Finance & Management, 59 - 88. Lending Club. (2015). Lending Club Statistics. Imechukuliwa toka Lending Club: https://www.lendingclub.com/info/download-data.action Pritchard, C. L. (2015). RiskManagement Concepts and Guidance . Boca Raton: CRC Press. Read More