A Pivot Study of an Innovative Approach to Identify Potential Buyers: A Case of Logistic Regression

Abstract

Purpose of this study is to demonstrate application of logistic regression as predictive tool when outcome variable is categorical. In real life, many times we come across with categorical outcome variable. These categorical variables are either binomial or multinomial. In such situations, we cannot use linear regression, as it requires metric criterion variable. Logistic regression is extension of linear regression. A case of antique shop is discussed in this paper. The shop owner was interested to identify prospects who would buy antiques. Case study approach has used in this paper. Paper presents important statistics and model fit criteria for logistic regression. Paper concludes with developing significant logistic model which predict potential buyer. This paper is useful to students, research scholars, academicians and business managers to understand application of logistic regression as predictive tool.