Some of the explanatory variables in a model uses a reference level when evaluating the effect driven by these variables. The reference level is needed when investigating the effect from a variable where the “absolute” value of that variable cannot be quantified by the model - and therefore a “relative” effect is quantified.
An example is the weather variable; the absolute impact of weather cannot be quantified by a model, since we have no observation of how your sales develop if there is no weather. Your sales are always affected by weather and therefore the models quantify the impact of changes in weather. Here the reference level comes in. The reference level is the period the recent development in effect, for instance for weather, is compared to.
Reference levels are used for the following variables in a model:
- Weather
- Macro
- Distribution
- Pricing
- Competitor pricing
- Branding
- Competitor branding
Was this article helpful?
That’s Great!
Thank you for your feedback
Sorry! We couldn't be helpful
Thank you for your feedback
Feedback sent
We appreciate your effort and will try to fix the article