To predict the sales of potentially weather-sensitive products (like umbrellas, bread and milk) around the time of major weather events (a weather event is any day in which more than an inch of rain or two inches of snow was observed) at different retail locations.
Problem Statement: To predict the sales of potentially weather-sensitive products (like umbrellas, bread and milk) around the time of major weather events (a weather event is any day in which more than an inch of rain or two inches of snow was observed) at dierent retail locations.
Problem Description: Supermarket stores like Walmart would like to know how much of each product is sold. It will help them to know how much of each product should be put in stock. This is especially true in times of extreme weather events. Intuitively, one can expect that the sales of umbrellas before a big thunderstorm would increase. But it’s difficult for replenishment managers to correctly predict the level of inventory needed to avoid being out-of-stock or overstock during and after that storm. This can be true for any other product that is potentially weather sensitive. We are predicting the sales of products using Walmart data for different retail locations. This project is a problem posted on Kaggle for Walmart.
Link to the Kaggle Problem: Store Sales Forecasting For Weather Sensitive Products
For entire details on the model creation and accuracy, please refer to the file final_report.pdf and for the implementation and code for this project, please refer to the code folder which has READ_ME.txt.
Contributors: Meghana Kotagiri(meghanakotagiri), G Neha (nehareddyg)