This project has moved. For the latest updates, please go here.

Pure Time Lag (Phase Shift) in Time Series Prediction

Oct 13, 2011 at 7:40 PM

Hello Everbody,

I have some time series models. I try to find k step ahead prediction of my model output. But for each model i always have a time lag (phase shift) between predicted data and real output. Predicted data looks like shadow of real data. When i increase k (prediction horizon) phase shift also increases. As i found it is a common problem in time series prediction, but unfortunately i could not find any way to overcome this problem.

Can anybody tell me a way or a trick to solve this problem?