When you want to practice your Data Science skills, there’s nothing better than getting your hand dirty with a Kaggle-style tabular data challenge.
In this post I give an overview of a project I did while at the Data Science Retreat Berlin together with my team mates Maryam Faramarzi and Naveen Korra.
This post contains a brief description of the analysis done and the final model.
You can find the full code for the project on GitHub. There you can also find notebooks with detailed explanations of the modeling.
The goal of this project was to predict the sales number of Rossman stores over time. The data was adapted from the Kaggle Rossman challenge. It includes the daily sales number of Rossman stores from January 2014 to July 2015.
In addition, information on the individual stores is provided, such as the type of store, the assortment of items sold, etc.