University of Twente Student Theses
An advice to improve the short-term return forecast
Maljaars, M (2020) An advice to improve the short-term return forecast.
PDF
3MB |
Abstract: | This research is conducted with the aim to improve the short-term return forecast of Bol.com. The company expanded significantly in the recent years and face difficulties with forecasting the associated number of returns. Several models are proposed to improve the current accuracy using return requests. Firstly, a model is developed to classify whether return requests will be true using Logistic Regression and Random Forest. Secondly, the timing for those truly classified returns is determined using LASSO, Poisson and Negative Binomial Regression. Based on those two complementary models, the total number of returns is predicted. In addition to a short-term return forecast on request level, a forecast on daily level is created using the aggregate of the return requests per day. |
Item Type: | Essay (Master) |
Clients: | Bol.com, Utrecht, Netherlands |
Faculty: | BMS: Behavioural, Management and Social Sciences |
Subject: | 01 general works, 30 exact sciences in general, 85 business administration, organizational science |
Programme: | Industrial Engineering and Management MSc (60029) |
Link to this item: | https://purl.utwente.nl/essays/82692 |
Export this item as: | BibTeX EndNote HTML Citation Reference Manager |
Repository Staff Only: item control page