Ｉ　ｓｅｅ．　Ｔｈａｎｋｓ　ａ　ｌｏｔ．

HI 
After getting my hands dirty, I am wondering if the online portfolio selection algorithms ( eg. Universal portfolio) are really a practical approach in real world   Have these methods been used in real trading? Or they are just for academical interest?

Hi 1. In White's reality check, what the smallest number n, prediction length, should be used to get sensible result?2. Can this method apply to adaptive schemes, like Kalman filer ?THKS & B RGDS

<t>Thanks acastaldo. Then the follow up is " If we find a profitable strategy in OU process, even by simulation, then we can, maybe hard, prove it profitable analytically" I.e. We can PROVE there is a real money machine ?My point is by assuming the spead is OU process, is it possible to find, by som...

<t>I am curious that is it correct to model pair-trading in O-U process. I.e. Modele the spread as O-U process since it's mean- revertingThe thing confusing me is the O-U process is martingale, so there is no profitable trading strategy ?!But people do model the spead in OU process and trade on that...

<t>Thanks Val,I think my model is different from American option, even it looks alike. The contingent event is moving average cross over. No price over some threshold.After moving average cross over, enter the market. If moving average cross down, exit the market. I am trying to find the analytical ...

<t>I am modeling a simple trading strategy in SDE terms:Price process: dS=u*S*dt+sig*S*dWP/L: Integral(I(M_day EMA -N_day EMA)dS(t), 0,T) where "I" means Indicator function. M<NMy question is how to deal with the stoch integral where integrand is an Indicator function of stopping time, like M_day EM...

Hi I am looking for the following paper on pair tradingFrazzo, J., Mougeot, N. & Smith, I (2002). Initiation of European pair trading research coverage. Lehman BrothersCould someone help? Thanks in advance.

Several months ago I saw Olsen was going to provide some HF data to Wilmott's reader. But now I cannot find here. Could some one help me?THKSray

Here is it.
http://www.princeton.edu/~bcf/calvet.pdf
You can also find relevant papers from same authors.

Any comment on multifractal process? It's said it's of no arbitrage opportunity (unlike FBM) and with long memory. Any cons?