Is it possible to compare in-sample and out-of-sample forecasts by calculating a MSPE for each? For example, my in sample period is say 12/1987-12/2015, and my out-of-sample forecasting period is 12/1993-12/2015 (I use an expanding window to forecast, with a burn-in period of 5 years to determine the initial parameters). Can I compare these two by taking the in-sample MSPE from 12/1993-12/2015, and compare that to out-of-sample? Or does this not even make sense? I want to see how my out-of-sample works compared to what my in-sample says, but I have no idea how this would look for an expanding window forecast, since it converges to the in-sample result eventually.


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.