Store Sales Time Series Forecasting

Forecasting Favorita Store Sales using Machine Learning

PROJECT CONTENTS

1. Introduction

2. Exploratory Data Analysis

3. Feature Engineering

4. Evaluation Metrics

5. Machine Learning Model Selection

6. Tune Hyperparameters

7. Exponentianal Moving Average

8. Conclusion

9. Summary

10. References

I. Introduction

Favorita Corporation develops commercial, industrial and real estate areas, present in Ecuador and 6 other countries in the region. They opened La Favorita supermarkets in country called Ecudor in South America which was the first self-service store in 1957.

In this project we’ll use time-series forecasting to forecast store sales on data from Corporación Favorita, a large Ecuadorian-based grocery retailer. Kaggle has published a knowledge based compitition to predict store sales by providing Favorita Corporation Stores data.
Learn more