Sales Prediction of Rossmann Store using Regression Algorithms
To come up with initial questions and to formulate our research strategy, we should first look at the given guidance from Rossmann itself:
Rossmann operates over 3,000 drug stores in 7 European countries. Currently, Rossmann store managers are tasked with predicting their daily sales for up to six weeks in advance. Store sales are influenced by many factors, including promotions, competition, school and state holidays, seasonality, and locality. With thousands of individual managers predicting sales based on their unique circumstances, the accuracy of results can be quite varied. In their first Kaggle competition, Rossmann is challenging you to predict 6 weeks of daily sales for 1,115 stores located across Germany. Reliable sales forecasts enable store managers to create effective staff schedules that increase productivity and motivation. By helping Rossmann create a robust prediction model, you will help store managers stay focused on what’s most important to them: their customers and their teams!
We will focus our Data Science project on the delivery of a robust statistical model, which is able to accurately predict a 6-week sales performance of Rossmann’s drugstores based on historical data By running through the data science process we will be able to answer the following research questions:
To what extend is sales performance influenced by factors like: promotions, competition, school and state holidays, seasonality, and locality. What is an appropriate model to predict sales?
Our approach follows the generic Data Science process. As we just did in the parts above, the process starts with asking an interesting question. With the case of Rossmann we have identified a case, which is interesting for us because the underlying issue of sales prediction is relevant for all kinds of businesses, too. The second step is to get the data. Then, we will explore our data, which is the usual step to take after obtaining data. We will show plots to illustrate properties, trends, anomalies and patterns of the data. After ploting the first results, we will apply different statistical models and compare them to choose the most performing one. Finally, we will answer our initial questions and discuss future research directions.