Abstract:

As the forecasting of a huge-size financial time series by training a standard SVM (Suppo~Vector Machine) will result in slow training speed and large memory spending,a decomposition-cooperation-weighted SVM regression is put forward and used to predict the stock index.In the proposed method,a large specimen set is de-composed into several subsets,and the SVMs in diferent subsets are independently extracted.According to the im-portance of the obtained SVMs,diferent error punishment degrees are obtained. Compared with the traditional SVM ,the propo sed method greatly speeds up the training process with almost the same precision.

Key words: support vector machine, decompo sition-cooperation-weighted support vector machine, time series, stock index