@Chins Harsh has been dabbling in mean reversion approach for a while now. Might be a good pointer to use his help to avoid some initial mistakes in your recent experiment.
