Problem: Determining the least-costly polling sample likely to give reliable results. You sell a seasonal product and you would like to find out how the seasons affect sales and your sales growth. This information would help you in forecasting sales and making future ordering decisions.
This model calculates seasonal factors that measure the effect each season has on sales and fits a straight line to the normalised data by minimising the squared error. The "Trend", or slope of the line, is the sales growth per season. The "Base" is the Sales figure at time zero. You have the sales data by season for
the last two years: |
|
