I am doing prediction in the Time Series context and I have a question about the

Question

I am doing prediction in the Time Series context and I have a question about the Cross-validation process. Working with the dataset, I first take the log transformation and then take the difference to remove the seasonal trend as well as linear trend. For CV, should I calculate the MSE based on the data after all the transformation or based on the data only after taking the log transformation? I am asking this is because they give me different rankings. Which method will give me the model that is able to generate better prediction?

Please answer this question ASAP!

Thank you!

Explanation / Answer

You were calculate the MSE for both the data i.e MSE of data after all the transformation also MSE of data after taking log transformation for prediction.

you have to consider data set which gives you minimum MSE.i.e which method gives you minimum MSE this method gives you approximate prediction .

Related Questions

Navigate

• Browse (All)
• Browse I
• Subjects

Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.