Financial Time Series Forecasting Using Directed-Weighted Chunking SVMs

Table 2

Performance comparison between the traditional SVMs and directed-weighted chunk SVMs.

SVM method	Sensitivity	Specificity	CPU time (ms)
Traditional SVMs	0.79	0.81	458231
Directed-weighted chunk SVMs.	0.83	0.85	65367

Dataset: IBM stock daily close prices in training data set (12075 data points, from December 31, 1999, up to December 31, 2007).
SVMs parameters: kernel is Gaussian function with 𝜎=0.02, 𝐶=8.5, 𝜀=0.001, and tolerance = 0.001.
Chunking methods: now we decomposed the training data into 1208 chunks by the time intervals of 10 days.
The CPU time covered the execution of entire algorithm excluding the file I/O time.