A General Framework for Enhancing Prediction Performance on Time Series Data
The code is written in R.
Total 3 baseline + 1 thesis methods:
- ARIMA
- HWES
- NNet
- GFEP
File Explanation:
<method>NSteps.R
(E.g. NNetNSteps.R): library code for specific methods.Features.R
: library code for extracting feature in my method.Exp<desc>.R
(E.g. ExpSVRNSteps.R): main code for experimenting. Above libs are called. Start from here.
Presentation slide: http://www.slideshare.net/chinhuichen/slide-print-fororal