Description:
Analysed customer purchase history dataset to identify significant variables affecting purchase patterns
Used Poisson Model and Poisson Regression to examine customer characteristics
Used NBD regression to difference in Expected and Observed values of total purchases made.
Key Results:
Significant variables affecting total number of purchases found : Income and hhzz i.e Household size
Wald Chi Square for variable hhzz and Income i.e 5.7661 and 5.8624 with associated p-value 0.0370 and 0.0680 shows that we could reject the null hypothesis and conclude that regression coefficient hhzz and Income are statistically different from zero in predicting the purchase patterns as compared to rest of the variables.
Tools used : SAS 9.4 and Excel 2013
Collaborators: Ishan Dindorkar, Rohan Ashok Patil, Valay Raval and Sridevi Allela
Rohan Ashok Patil 