Predicting Impulsive Buyers: A Comparative Study of Binary Classifiers' Discriminative Ability

Researches over the years have posited that most of the purchasing decisions made by shoppers are unplanned and mostly made inside the store. Every year retailers spend huge amount on in-store promotion and store environment to gain consumers' attention and stimulate them for such impulsive purchasing. The dominant impact of unplanned purchasing behaviour advocates its comprehensive understanding in retail business. Inability of marketers to accurately forecast this unplanned shopping leads to situations of either over-stocking or under-stocking of inventories, causing increased cost of holding inventory or huge loss of customers respectively. In this research, accuracy in predicting impulse buying behavior using six modeling techniques, utilizing R-3.2.1 as a statistical tool, is tested with data drawn from two Indian metropolitan cities. The main objective is to analyze the relationship between independent variables (influencing factors) and impulsive buying, and comparing the effectiveness of these six modelling techniques. The discriminative ability evidences that forecasting performance of logistic regression is superior to all other techniques in terms of predicting power. The findings of the study offer a number of implications for retailers and marketers. The managerial implications of the study along with scope of further research have been addressed.