Skip to content
No description, website, or topics provided.
Jupyter Notebook
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
Presentation_and_Results
ARIMA_Model_Pipeline.ipynb
README.md
SARIMA_Model_Pipeline.ipynb
Sales_Forecasting_EDA.ipynb

README.md

Forecasting Retail Store Demand with Time-Series Analysis

At one point or another we’ve all had a shopping experience in which we’ve gone out with the intention of buying a specific item, only to find that the item we had been looking for was out-of-stock. This is a common frustration that most people can relate to and as such I wanted to investigate whether I could solve this problem using data science.

The objective of this analysis was to forecast store-level demand across various retail locations to ultimately inform business decisions relating to the prediction of short-term and long-term store performance, demand and financial planning, as well as inventory management and supply decisions. In order to address these business problems store-level data was collected from the online machine learning community Kaggle which provided five years of sales data across ten different store locations.

Several different time-series models for each store location were tested including autoregressive (AR), autoregressive moving average (ARMA), autoregressive integrated moving average (ARIMA), as well as a seasonal ARIMA (SARIMA) model which proved to be the best performing model for retail sales data. The models evaluated for this analysis may ultimately be leveraged to help predict future sales and plan for fluctuations in store demand across various retail locations.

You can’t perform that action at this time.