If I wanted to forecast sales in a grocery store and I knew factors that would influence grocery sales, would there be a reason why I should use a multivariate time series model vs. time series models with the influential factors as exogenous? Like, would there be some type of possible performance benefit in building a multivariate model vs. time series model with exogenous factors?

Edit: I guess if you don't have forecasts of the indicators then using these indicators as exogenous factors would not be possible, but does this mean that multivariate forecast, could result in a better overall forecast of grocery sales?



Your Answer

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