The Application of Random Models in Forecasting Commercial Bank's Saving Deposits Customer Purchase Behavior

摘要

In this paper, we use the random theory into analyzing commercial bank's saving deposits customer future purchase behavior. We build the purchase times and amount models according to the characteristics of commercial bank's products based on redefining these variables and modifying the models in retail industry, and then use the actual transaction data of one commercial bank's saving deposits to validate these models. In addition, we compare the two models of purchase times, Pareto/NBD and BG/NBD model, and come to a conclusion that we can use the BG/NBD model to take place the Pareto /NBD model when the time and cost is limited. We use the gamma-gamma model to analyze purchase amount, and further calculate the future transaction volume and profit It's meaningful to provide numeric decision-making for managers to guidance of customer relationship management.