Another post for the last image (FF 5-image limit):
http://dl.dropbox.com/u/190212/ff/quant_ema_6.png
http://dl.dropbox.com/u/190212/ff/quant_ema_6.png
Trade another day through Systematic Trading 203 replies
Nonparametric Association Measures in Systematic Trading 57 replies
Anyone use Systematic from Alpari? 1 reply
A Systematic Approach to Markets 68 replies
Systematic Trading vs. Discretionary Trading 14 replies
DislikedHmm, what do you mean by a single trend? HHT is a frequency domain analysis, it will output multiple trends (cycles). So you can have a big trend down, and a smaller trend up.Ignored
DislikedThere are lots of resources about portfolio construction, one starting point could be to read about modern/post-modern portfolio theory and continue from there.Ignored
Disliked.. due to the high frequency of trades and short holding periods (lower risk).Ignored
DislikedThere was something BlackMage wrote which struck me:
I'm not sure if I understood him correctly: short holding periods (lower risk)
Hi guys, what are your thoughts on this:
is a shorter holding period indeed lower risk?
Ignored
DislikedI recall studying modern portfolio theory in college and to be blunt, I wasn't terribly impressed. I tend to dive into the deep end on this stuff, so maybe I asked the wrong initial question. Have you got any interesting scholarly papers on the subject of portfolio switching?Ignored
DislikedI've read a paper which defined in mathematical form basic TA patterns, and then proceeded to backtest them on daily equity data. The only one they found to have a significant statistical significance was the H&S pattern, but it was not very profitable.Ignored
QuoteDislikedLike I said in another thread, I've not seen a single candlestick chart in not even one of the quant papers I have. All use continuous data
DislikedI wish you guys best of luck. I'll stop posting here since it seems nobody is really interested in sharing stuff that really works, as posted by Craig and mikkom latelyIgnored
DislikedIf I would give something away from my algo for exmaple it would be about the basic principles (and I think I have told these already) - that it's based on principle of evolution like most of the stuff I work with because that's what I know best. Also I think that portfolio combination is all about what the trader is personally looking for in the results and surprisingly many people haven't quantified this in meaningful manner. After you know what you are looking for, you can easily find combinations of strategies that fit to your requirements...Ignored
DislikedCraig, I was going to ask this before but I forgot - do you use any of the same algos in all the different markets (fx/fut/equities) or do you have specialized versions for markets/clusters of instruments?Ignored
DislikedHere is some food for thought, not directly relevant but shows the power of different combinations.
(source: http://efjournal.com/blog/2009/jan_mar_2009.html)
There are some studies about this in real market situations.
(I'm not using any derivation of this in my algos [as far as I know]).
http://efjournal.com/images/blog/parrondo.gifIgnored
DislikedThis is an interesting subject, there was a big thread on ET on this subject (http://www.elitetrader.com/vb/showth...rrondo+paradox).Ignored
DislikedAppreciate the post. This is actually what I was attempting to get at - I admit I'm a bit of a johnny come lately to this thread and can understand the desire for secrecy - I tend to also be private on the boards regarding things that have taken me years of blood sweat and tears to figure out. Discussion of general principles is exactly where I'm coming from and was the intent of my previous post, not to elicit explicitly what you are doing.
Before reading the post I'm replying to above, I was pondering on your reply to my post I and basically...Ignored
Disliked
Guess what happen when I combine them together? The annualized daily sharpe ratio of the portfolio surge to 6.5 and the profit factor double up to 3.0! The relative drawdown also reduced significantly more than I expected!Ignored
DislikedForexQuant,
I can understand that the relative drawdown is less. If one system goes up and the other goes down this would have a positive effect (if every individual system has a expectancy of >50% win). What I don't understand is how the profit factor can double? The individual losses and profits of each system remain the same so how can it double the profit factor to 3 (baring in mind each individual system has pf <2)?Ignored
DislikedI read a rather interesting post from another FF member (Ronald Raygun) where he broke down his strategies into categories: Trend, counter trend, and breakout. He mentioned that he has around 30 total strategies split between the 3 categories that he trades continuously and just uses his indication of which leg of the cycle is dominant to switch sizing amongst the strategies.
On that topic, do you subscribe to this idea that trading different styles of systems together can help reduce drawdown and improve total return? So far I've tested this...Ignored