[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[EquisMetaStock Group] SPY Trading System Back Testing Questions



PureBytes Links

Trading Reference Links

This was backtested using TradeStation.

I took the Tradestation results and did what I suggested all of you 
do, I set it up on chart; I manually marked the entry and exit 
signals; and then I figured out the trade profit and losses by hand. 
When you do that your results will vary from the backtest results 
slightly but you then know how well you will do trading the sytem and 
not the computer. The computer won't make the trades for you, so you 
might as well figure out how much less efficient you are than the 
computer. That way you won't get surprised in live trading. 

I do not use mechanical test results from any program without the 
hand anaylsis. When I backtest with a mechanical system I only 
consider the results to be relative--that is comparable from one to 
another on the same testing platform. However, they have nothing to 
do with live trading. Often a mechanical system will give you great 
results until you go to trade it and you can't read the entry and 
exit signals, or process the information fast enough because we're 
human and not a computers. 

This system has already been backtested by a lot of people. You don't 
need to reinvent the wheel. If you look at three or four years worth 
of data and pick out the signals by eyeball, it will tell you 
everything you need to know. 

As I explained, I study and tinker with systems development, and this 
came from some of my books. I've put this together based on the work 
of many other people who are very competent. 

If you backtest and get results other than the ones I've told you 
about, I can't explain to you why you got different results. I don't 
have clue as to what you have done. If you set the system up right, 
follow the trading rules, eyeball test it, anc calculate the results 
properly by hand, you will get very similiar results to the 
tradestation results. 

If you backtest this using other systems, who knows what you'll get. 
I certainly don't know. It doesn't matter. If you set up the 
Geometric Index and test it using the same platform with the same 
setup, it will give lower results than the Arithmetic Index, which is 
what Tradestation said. While your comparison results will be the 
same, your P&L numbers will be different.  

I'm sorry, but I don't have time to answer any more questions on 
backtesting. Anyone who wants to use the system can. If you want to 
do 10,000,000 backtests on it yourself, be my guest. I just can't 
spend the time helping you figure out how your backtesting works, 
what parameters to use, what assumptions to make, etc. Backtesting is 
a black hole for time, energy and misleading results. 

This system only predicts market moves as a whole, so it only works 
with the market as a whole, thus SPY. 

I was asked if the system was curve fitted! How can you curve fit a 
4% price move. It either moves 4% or it doesn't! I was asked if it 
was tested on out of sample data! That's only relevant if the 
equations can be curve fitted. 

Sorry no more backtesting questions, please.

Good luck! 

JO



------------------------ Yahoo! Groups Sponsor ---------------------~-->
Buy Ink Cartridges or Refill Kits for your HP, Epson, Canon or Lexmark
Printer at MyInks.com. Free s/h on orders $50 or more to the US & Canada.
http://www.c1tracking.com/l.asp?cid=5511
http://us.click.yahoo.com/mOAaAA/3exGAA/qnsNAA/BefplB/TM
---------------------------------------------------------------------~->

To unsubscribe from this group, send an email to:
equismetastock-unsubscribe@xxxxxxxxxxxxxxx

 

Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/