In the original post I said there was a drag because the discussion was on a closed set of trades. Within that set there certainly is a drag effect. In a followup discussion it appeared that my statement had been used as a universal statement about drag, which was untrue. From a practical standpoint I use drag as the norm and the no-drag by virtue of a string of winners as a bonus. Anyone telling their boss that someday our drag effect will be wiped out by a string of x number of consecutive winners would be laughed off the desk. It's better to treat a large winning streak as a bonus rather than to expect it as part of a distribution of trades. To treat it any other way is to trade as a engineer rather than as a trader. The title of the thread is 100% accurate since I've posted material at various stages of my research. My current work looks very different from what I've posted in the past as I've found new insights. I've turned off my pm's to avoid getting the usual 10-20 queries. I've chosen to skip posting new material on ET as I don't want it in the public domain at this time. Good luck to all those doing systematic research.
Acrary, I have but one last question. In your system development thread you used three models where you explained importance of correlation. You said the first model had an edge and the other two had no edge. Nevertheless, how do you come up with three models that trade the same market with such low low correlations? -0.026, -0.099 and -0.01
I know this will sound redundant, but if you had classified all the different market conditions you would see the answer. If model A is trend following then look for conditions for model B where countertrend trading is profitable. If A and B exist how about a range bound trading model for model C. Then how about a reversal model for model D. Once the data is properly classified you can find all these models and more.
Thanks for the quick answer If you don't mind.. do you use daily or intraday data to develop/test range bound model? I can hardly imagine testing range bound 1-day holding period model with daily data spanning over many years.
Ok I guess I'll have to spend the next few years trying to understand how to properly classificate market behaviors. I wonder if Toby Crabel's analysis methods are of any use in trying to analyse intraday price action and isolating conditions (price patterns) where those behaviors happen.
acrary taught people how to think in a systematic fashion, which is the beginning of trading for your livlihood and mastering yourself, which actually is what is required to trade successfully, not mastering the markets, as many think. let's put it this way, I have several threads archived under "Lessons from the Professors" ... and acrary is definitely on that list. it's amazing some of the knowledge floating around this place, while people try to denigrate it and vendors run around on their sim machines talking about their holy grails ... it's a frick'in zoo.