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

[amibroker] Statistical relevance of backtesting single issues or indexes



PureBytes Links

Trading Reference Links

Meta-Stock, TC-2000, TradeStation & AIQ are all software packages that
advertise their ability to perform back test functionality on stocks.
The trouble is that all of them restrict back testing to the currently
displayed symbol.

Over the years, I've looked at them all and rejected them because I
believe that back testing individual equities or composite indexes is
statistically insignificant.  And in AIQ's case, their system was
composed of `black box' technology. (I'd just as soon toss a few
chicken bones into a circle and then try to divine mkt direction from
the bone layout configurations.)  Maybe AIQ is different now, but I
rarely waste my time taking second looks at flawed approaches.

It was always my belief that any viable mechanical trading system must
prove itself over multiple tradable securities… not single indexes or
issues. AND that those trading system results must be pretty
consistent over varying time frames. (That does not mean that you
don't experience draw downs, but draw downs easily fit within an up
trending equity channel.) 

I NEVER bet money on trading systems that are restricted to single
issues or indexes. But rather I prefer strategies which have been
successfully back tested against a statistically significant number of
equities covering a variety of time periods. And with LARGE numbers of
back tested trades compounding of trade size is usually impractical. I
have always used fixed trade size. If you have a small account then
this approach is unappealing. But if your account size is a bit
bigger, then fixed trade size is a steady growth strategy.

Amibroker is the only `off the shelf' tool I am aware that allows you
to take the next step and validate trading results using combinations
of securities over varying time frames.

My point is that if you think you have a good trading system that
works on the NDX or the QQQ, and then expand the back test to groups
including sets and subsets of NYSE, AMEX and NASD stocks to help
assess the predictability and stability of the strategy.

Just my opinion.

Phsst



Post AmiQuote-related messages ONLY to: amiquote@xxxxxxxxxxxxxxx 
(Web page: http://groups.yahoo.com/group/amiquote/messages/)

Check group FAQ at: http://groups.yahoo.com/group/amibroker/files/groupfaq.html 

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