I don't think that anybody would share that knowledge. One would propably make some money already. However, if you really want to do some analysis it maybe helpful to concentrate on evaluation of risk or risk management. There exist other more realistic prediction models on other domains. Look for example in weather forecast research.
i know many different paradigm applied to this field such as :ANN, Case-based reasoning,Back-propagation neural networks,Support vector machines or hybrid of many method but i want some guide to give me direction for example some review paper or some one work at this topic for many years
i want predict in market and i look for best way but i can't find proper answer.I think a lot of different method apply to this area but which of them is better and why ? is there any good review paper ?and how better i search or ask this question?