Frequency Model of Credit Payment using Bayesian Geometric Regression and Bayesian Mixture Geometric Regression

In distributing funds to customers as credit, multi-finance companies have two necessary risks, i.e. prepayment risk, and default risk. The default risk can be minimized by determining the factors that affect the survival of customers to make credit payment, in terms of frequency of credit payments by customers that are distributed geometry. The proposed modelling is using Bayesian Geometric Regression and Bayesian Mixture Geometric Regression. The best model of this research is modelling using Bayesian Geometric Regression method because it has lower DIC values than Bayesian Mixture Geometric Regression. Modelling using Bayesian Geometric Regression show the significant variables are marital status, down payment, installment length, length of stay, and insurance.

Mortgage Dollar Roll

Mortgage dollar roll, the most common financing strategy for agency MBS, differs from repo in that the returned collateral can differ from those received. Also, MBS ownership changes hands in the funding period. We show that dollar roll “specialness,” how much implied financing rates fall below MBS repo rates, (1) increases in the value of the cheapest-to-deliver option, (2) decreases in the leverage of primary dealers, (3) decreases in prepayment risk exposure during the financing period, and (4) decreases in MBS returns. The Federal Reserve’s dollar roll sales in quantitative easing operations are associated with lower specialness. Received February 3, 2016; editorial decision July 30, 2018 by Editor Itay Goldstein. Authors have furnished an Internet Appendix, which is available on the Oxford University Press Web site next to the link to the final published paper online.