Tid: 14 september 2018 kl 09.15-10.00.

Föredragshållare: Sofia Karlsson

Title: Purchase behaviour analysis in the retail industry using Generalized Linear Models

Abstract: This master thesis uses applied mathematical statistics to analyse purchase behaviour based on customer data of the Swedish brand Indiska. The aim of the study is to build a model that can help predicting the sales quantities of different product classes and identify which factors are the most significant in the different models and furthermore, to create an algorithm that can provide suggested product combinations in the purchasing process. Generalized linear models with a Negative binomial distribution are applied to retrieve the predicted sales quantity. Moreover, conditional probability is used in the algorithm which results in a product recommendation engine based on the calculated conditional probability that the suggested combinations are purchased. From the findings, it can be concluded that all variables considered in the models; original price, purchase month, colour, cluster, purchase country and channel are significant for the predicted outcome of the sales quantity for each product class. Furthermore, by using conditional probability and historical sales data, an algorithm can be constructed which creates recommendations of product combinations of either one or two products that can be bought together with an initial product that a customer shows interest in.

The full report (pdf)

Till seminarielistan
To the list of seminars