This is the R (Microsoft R Server) code for Retail Forecasting template using SQL Server R Services.
Two modeling solutions are provided for this template :
- Time Series Forecasting:
- Seasonal Trend Decomposition using Loess (STL) + Exponential Smoothing (ETS);
- Seasonal Naive;
- Seasonal Trend Decomposition using Loess + AutoRegressive Integrated Moving Average (ARIMA).
- Regression Models:
- Boosted Decision Tree Regression
- Random Forest Regression
It consists of the following files:
| File | Description |
|---|---|
| 01-data-preprocess.R | Load data to SQL tables, apply business rules to the data, filling NAs, etc |
| 02-time-series-modeling | Train the time series models |
| 03-feature-engineering | Feature engineering for regression models |
| 04-regression-modeling | Train the regression models> |
| 05-evaluate-models | Evaluate the regression models |
A detailed description of the template, implemented in Azure Machine Learning Studio can be found here.