I am trying to predict category sales for automotive market. The reason I need this variable is because I have been using it in my scoring data-set for a marketing mix model (time series regression model).

In past I have simply used moving average and simple time series mean and median forecasting. But now I need more concrete methodology for prediction as during actualization (scoring after I have the actual category sales values) I have been noticing a lot of deviation.

I have historical category sales from 2013 monthly and I was thinking of downloading data like gas prices, inflation rate, GDP and median income to build a separate regression model for the category sales. Can anyone direct me or suggest a better way.

Thank You

  • $\begingroup$ Make sure your "regression" approach correctly takes into account "time series complications,consequences & opportunities" See the following web-gem autobox.com/pdfs/regvsbox-old.pdf for some pointers. $\endgroup$ – IrishStat Jul 23 '19 at 15:07

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.