Page 1 of 1
Posted: Tue Aug 31, 2004 3:09 pm
In my research, i have found that shorting stocks, using a long-term trend following system, doesn't work very well. However, this could very well be because stocks that decline and get delisted no longer appear in traditional data sources. On the other hand it could be that because the long term trend of the market is up one shouldn't fight the market. Also, long positions have the potential for unlimited gains with limited risk, and shorts have the potential for limited gains with unlimited risk. I don't know whether my time should be spent looking for stocks that will hold their value during broad market downturns or instead, looking for short candidates. What has your research shown?
Posted: Tue Aug 31, 2004 8:35 pm
Veritrader comes with stock data for the Nasdaq-100 stocks, starting in 1994 and going to 2004. Running some systems on that data, here's the best I've seen so far. It isn't wonderful; system equity had a tremendous zoom-spike-peak-collapse in 2000, and the peak hasn't yet been surpassed. Therefore the system is in "drawdown" and has been for quite some time.
I'm telling the software to risk a maximum of 0.2% of equity per trade, which means that for a 100-stock portfolio, margin is definitely being used. I plugged in the Veritrader standard 6% for margin interest rate. Stop loss orders are placed at 6 ATR's (20 day ATR) beyond the entry fill. As you can see from the attached figure, the performance on the Short side is dismal.
Is it "long term" trendfollowing? I think so. There's 1325 trades in 10 years, with a portfolio of 100 stocks. About 1.3 trades per year in each stock, which I would call long term.
Posted: Tue Aug 31, 2004 11:48 pm
What is your entry/exit criteria?
Posted: Wed Sep 01, 2004 7:49 am
I think it might be appropriate levijean for you to post your system rules and your quantitative research results including details like compound annual growth rate, maximum drawdown depth, maximum drawdown duration, portfolio size, and so forth.
Posted: Wed Sep 01, 2004 9:12 am
I'm still working on it, and as of today, its nothing to write home about.
Posted: Wed Sep 01, 2004 11:21 am
Does the given Nasdaq 100 list properly account for membership changes?
Also, I think the Nasdaq is too tech heavy. Would be nice to see the results on SP 500, Russell 1000, or (even better) Russell 3000.
But then, of course, the results might be skewed if membership changes are not handled properly.
P.S. - if anyone knows where to get historical membership listings for the Russell indices (free or cheaply...) pre-2001, I would like to know.
Posted: Fri Sep 10, 2004 7:47 am
I am a french stock trader and I am currently looking for database of historical charts going back as far as possible in the past. The paris stock exchange does not provide any information of that kind and I am sure it exists.
Does anybody know where I could find such information?
Sorry for my approximative english!