Gerald P. Ifurung

04/11/2011

Keller School of Management

Executive Summary

Every portfolio has a set of delinquent customers who do not make their payments on time. The financial institution has to undertake collection activities on these customers to recover the amounts due. A lot of collection resources are wasted on customers who are difficult or impossible to recover. Predictive analytics can help optimize the allocation of collection resources by identifying the most effective collection agencies, contact strategies, legal actions and other strategies to each customer, thus significantly increasing recovery at the same time reducing collection costs. A random sample of accounts closed out during the month of January through June will be used in determining if the size of the bill has an effect on the number of days the bill is late. The statistical analysis of the data involves the application of regression analysis. Based on the calculated value of correlation coefficient, there is no relationship between the size of the bill and the number of days to collect. .

Introduction

The author was hired by the Quick Stab Collection Agency (QSCA) on a contractual basis to assist the company in auditing potential business in buying the rights to collect debts from its original owners. QSCA is a collection agency that specializes in very profitable small accounts and avoids risky collections. Profitability at QSCA depends critically on the number of days to collect the payment and the amount of bill, as well as on the discount rate offered. A random sample of accounts closed out during the month of January through June will be used in determining if the size of the bill has an effect on the number of days the bill is late. The statistical analysis of the data will involve the application of regression analysis. The analysis will determine if the size of the bill relate to the number of days of late payment and whether there exist a correlation if the customer is either a residential or commercial entity.

Analysis and Methods Section

Regression Analysis| | | | | | |

| | | | | | | |

| r² | 0.002 | n | 96 | | | |

| r | -0.042 | k | 1 | | | |

| Std. Error | 23.724 | Dep. Var. | DAYS| | | |

| | | | | | | |

ANOVA table| | | | | | | |

Source| SS | df | MS| F| p-value| | |

Regression| 91.9166 | 1 | 91.9166 | 0.16| .6870| | | Residual| 52,904.4896 | 94 | 562.8137 | | | | |

Total| 52,996.4063 | 95 | | | | | |

| | | | | | | |

| | | | | | | |

Regression output| | | | confidence interval| |

variables| coefficients| std. error | t (df=94)| p-value| 95% lower| 95% upper| | Intercept| 51.9839 | 5.9641 | 8.716 | 9.84E-14| 40.1421 | 63.8257 | | BILL| -0.0126 | 0.0313 | -0.404 | .6870| -0.0747 | 0.0495 | |

According to B. Weaver (Chap 4, P. 7), “Correlation is concerned with the relationship between two or more variables. Regression is typically concerned with the using the relationship for prediction. Correlation, on the other hand, is concerned with finding out whether there is a relationship, and if there is, determining its strength and direction. If two variables are correlated, then one variable may be the cause of the other. If two variables are not correlated, there is no causal relationship between the two or at least not a linear causal relationship.” From the analysis, the following statistical interpretations are summarized from the above data set: * Regression Equation: y = 51.9839 – 0.0126x. Interpretation of b0 : The estimated number of days to collect the payment is 52 days. . Interpretation of b1: For each additional dollar amount of overdue bill, the company can expect an increase of 0.012 days in late payment. * Standard...