Sp500 test results comments ?

[jsd]

hmm right ive come up with this mechanical on the sp500 cash 1995/2001 six years data need the 2001 to present to see how it coped.. EOD ASC II. txt format . left entry inputs the same and broke the 6 year series down into 3 lots of 2 years to highlight any potential over optimization, as per schwager on futures thinking..

95/97 however i did adjust the money stops to account for the smaller daily range at that time in the index. 97/99 and 99/01 money stops bat off the same levels though . all entry /exit inputs throughout all 3 series constant.

looking for comments from more experienced alchemists for negs and positives.

any pointers re last 4 years ASCII.txt for the SP500 availabilty appreciated..

trades long and short mostly invested all the time. shortside while not great in the uptrend held its own. opionions wanted really..


EOD data cash index
no costs factored in
all values in Full SP500 index points.

jd
hope the files attach ok..

 

[JonnyT]

For an EOD system this is perhaps the poorest I have ever seen published.

Sorry for doing a Simon Cowell, but a few hundred dollars profit over a two year period over multiple trades AND without costs etc...

JonnyT

 

[juanbyte]

If your net profit figures are full points and not dollars I think it is quite good especially when most people actuall lose.

The low number of trades means your costs will be negligable.

 

[jsd]

Hi Jonny the few hundred dollars are Full index points($250.00) ie 1999/2001 = 511.77 =$127,942.50

 

[sidinuk]

You need to test the system on the instrument you intend to trade rather than the cash index. There could well be a few stops hit out of hours which eod cash index wouldn't pick up because it is closed.

Other than that it's very difficult to say much without knowing what the system rules are.

 

[fowkesp]

Agree with sidinuk. You will need to test with a tradeable instrument, rather than SP Cash. Also, if your entries are anywhere around the market open, the results will be seriously awry. The Cash and Futures prices (or other tradeable) go way out of line during the first few minutes, so any strategy relying on Cash prices around the opening time will get false results.

Apart from that, could you really trade a system with the drawdowns shown in your stats? Might be an idea to check you haven't got a rogue trade or two in there that has screwed the results.

Paul

 

[jsd]

hi thanks for the feed back,

I hear what your saying about trading off of the cash, initial trades entered have a 12 full point money stop, assume that this will hold most overnight skews on globex, its either then stopped out on the money or the system exit criterior.

I will test it on actual futures but initially used the cash to see if there was continuity within a chunk of data... I'll check im sure ive got the corresponding futures data on cd upto 2001 is it better (or more realistic) to test it on continuous futures or the individual shorter contracts...? i'll do both and see... enters on the close of bar, not the next open, early days at the moment.. havent checked limits or stop orders on entry either.

Drawdowns now whats considered ball park ? subjective I suppose relative to return..

jd