Long-term prediction of time series using a parsimonious set of inputs and LS-SVM
Jarkko Tikka and Jaakko Hollmen
In: European Symposium on Time Series Prediction (ESTSP 2007), 7 Feb - 9 Feb 2007, Espoo, Finland.
Time series prediction is an important problem in many areas
of science and engineering. We investigate the use of a parsimonious set of
autoregressive variables in the long-term prediction task using the direct
prediction approach. We use a fast input selection algorithm on a large set
of autoregressive variables for different direct predictors, and train non-
linear models (LS-SVM and a committee of LS-SVM) on the parsimonious
set of non-contiguous set of autoregressive variables. Results will be shown
for the time series competition task.