Skip to content

This is a retail case study. Help a major retailer forecast their sales to minimize revenue loss from the unavailability of products by investing accordingly.

Notifications You must be signed in to change notification settings

myinnos/Retail-Case-Study-Kaggle

Repository files navigation

Retail Case Study

This is a retail case study. Help a major retailer forecast their sales!

Forecasting is an important approach to plan the future effectively and efficiently. A leading retailer in USA, wants to forecast sales for their product categories in their store based on the sales history of each category. Sales forecast has very high influence on the performance of the company’s business and hence these sales forecasts can be used to estimate company’s success or performance in the coming year. Accurate forecasts may lead to better decisions in business. Sales or revenues forecasting is very important for retail operations. Forecasting of retail sales helps retailer to take necessary measures to plan their budgets or investments in a period (monthly, yearly) among different product categories like women clothing, men clothing and other clothing and at the same time they can plan to minimize revenue loss from unavailability of products by investing accordingly.

You need to help predict the RMSE values for the test set - there are 4 datasets 1. Macro Economic Dataset 2. Events and Holidays Dataset 3. Weather Data Set 4. Train Data (Sales and the Year/Month)

Additionally there is the test data set that you need to prepare the results.

Data Description

Retail Sales Forecasting: 1. Sales data: “Train.csv” This table contains the Temporal data like Year, Month, and Product category and Sales (In ThousandDollars), provided for the period 2009 to 2014

  1. Macro Economic Data: “MacroEconomicData.xlsx” This data is provided for the period from 2009 to 2016 with the details like CPI, GDP, Cotton production, mill usage, unemployment rate etc.
  2. Weather Data
  3. Events and Holiday Data
  4. Attributes Details: “AttributesDescription.xlsx” This has the details of attributes for the datasets cited above (1 to 4)

The objective is to forecast the 'Sales(In ThousandDollars)' column (target variable). The error metric is the RMSE (Root Mean Squared Error).

Test data is in sequence for the 2015 month wise and product wise (so totally 36 columns)

TestData

TrainData

Results

Results

About

This is a retail case study. Help a major retailer forecast their sales to minimize revenue loss from the unavailability of products by investing accordingly.

Resources

Stars

Watchers

Forks