After figuring out some of the SVM parameters, I did a comparison of an MLP (feedforward neural network) technique vs. the SVM (support vector regression) technique for use as a predictor. The data were split into train/test set at a ratio of 9/1, both the SVM and the MLP were trained with those data and this was repeated a few (20) times. It turns out that the neural network seems to perform better and oscillates less over the trial runs. The following figures tell the tale more precisely:

Mean Absolute Error, MLP vs. SVM

Root Mean Squared Error, MLP vs. SVM

As usual, the matlab script, (also linked on the left). I did have some issues first with the nntraintool call in the latest 6.0 version of the Neural Network toolbox. Those issues were the same as those in this post at Matlab Central. I solved them by erasing my stored matlab path variable, resetting it to its defaults and adding the necessary paths to my own scripts later. Worked out fine.