Forecasting the product return rate in the e-commerce context

Abstract

The main objective of this thesis is to understand and analyze the impact of product returns on the losses of e-commerce companies. To this end, data from an e-commerce company in the fashion sector was used. The variables analyzed were: product return rate, company’s losses, and company’s profit. Studying the relationship between these three variables revealed, on the one hand, that the higher the product return rate, i.e., the percentage of products returned, the higher the company’s losses, i.e., the costs the company incurs when its customers return the product they bought. On the other hand, the higher the company’s losses, the lower the profits. Therefore, there is a significant impact and a relationship between returned products and company losses. As a result, e-commerce companies should be concerned and carry out periodic monitoring, as well as develop strategies to reduce the percentage of returned products to reduce company losses. One way of doing this is by using forecasting methods. With this in mind, I studied and analyzed four forecasting models: HP filter, MED filter, AR(1) model, and Theta Model. Among all these models, the best-performing model was the Theta Model, regardless of the variable under study. Therefore, this model was applied to make the monthly forecast for next year. According to the forecast values, we concluded that if the company doesn’t implement any strategy to ensure a reduction in returned products, then the company’s profits will fall.

Date of Award	25 Jan 2024
Original language	English
Awarding Institution	Universidade Católica Portuguesa
Supervisor	Pedro Afonso Fernandes (Supervisor)