Simplified Methods of fitting the truncated Negative Binomial Distribution: A model that allows for Non users

Abstract

Retailers monitor customer buying-behaviour as a measure of their stores’ success. However, summary measures such as the total buying-behaviour provides little insight about individual-level shopping behaviour. Additionally, behaviour may evolve over time, especially in a changing environment like the Internet.

This research developed a useful stochastic model for analysing period to period fluctuations in sales thereby generalizing the model proposed by Goodhardt and Ehrenberg to allow for nonbuyers of the product category. So as the composition of the customer population changes (e.g., as customers mature or as large numbers of new and inexperienced Internet shoppers enter the market), the overall degree of buyer heterogeneity that each store faces may change.

A systematic bias in their simple negative binomial distribution [NBD] model is demonstrated. If the proportion of nonbuyers is large, the simple model will be wrong. As a result, frequent buyers often comprise the preferred target segment. We find evidence supporting the fact that people who visit a store more frequently are more likely to buy. We also gives explicit formula and directions that allow a moderately analyst to perform his own conditional trend analysis.

KEYWORDS:  Buying Behaviour, Negative Binomial Distribution,  Heterogeneity, Conditional Trend Analysis.