System testing questions.

jim c · Jul 15, 2009

Quote from nitro:

Jim,

Curve fitting has a strong correlation to number of parameters. Most system traders aren't aware of all the hidden parameters that a system has. For example, using a particular date range to test a system against is a hidden variable. Or how about which symbol to test a system against is a hidden variable. Basically, any decision that you make is potentially a variable, that while not necessarily explicit in your actual system code, nevertheless exists as a constraint on it's performance, and therefore could be selecting a curve fit. The act of not having stops is clearly a hidden variable.

If you said something like,

Code:

VIXVALUE = 15 if VIX < VIXVALUE then no stops else stops = ATR(oversomeperiod)

That would make the variable explicit, and should probably be done for anyone that doesn't have infinite resouces.
More...
This makes sense. In my simple mind I imagine "no stops" as the simplest or purest way to backtest (even though I would NEVER trade without stops). I guess I need to think about this for a while.
I sort of asked the question in the begining of the thread...would trading just the profitable stocks be curve fitting also? Im thinking it would. Anyway thx for the food for thought. Again, I could never trade without stops but here is Ernie Chans blog talking a little about it. http://epchan.blogspot.com/ thx again guys! JIm

TrueStory · Jul 22, 2009

Quote from jim c:

I apprectiate the response. This is just a rough way to see of the system would work at all. I would like to think that using stops with this system would actually improve it. ( I deff think it would, esp after looking at the MAE) Maybe you could talk a little more about how you think using no stops would be considered curve fitting. Again, I really apprectiate the response! Jim
More...

Well, in a manner of speaking, nitro said it for me. But the most important point is that trading without a stop is not possible without unlimited resources. That's all that really matters. It HAS to be part of the system, hence why I said that without it, you don't have a system at all.

Technician · Jul 30, 2009

Quote from nitro:

Add the following filter:

1) If VIX < 20, then do as you say. If VIX > 20, then only do what you say if market opens higher than FV.
More...

Can you explain how this works? I couldn't figure it out.

alpri16 · Jul 31, 2009

Jim:

In your original post your spread sheet shows that the most trades taken in any single issue is 48 over a period of 5 years. Sounds like a very selective set up. Curious, what are the parameters for the short action?

Paul

jim c · Aug 3, 2009

Quote from alpri16:

Jim:

In your original post your spread sheet shows that the most trades taken in any single issue is 48 over a period of 5 years. Sounds like a very selective set up. Curious, what are the parameters for the short action?

Paul
More...

Its a huge deep dark secret. If stock is less than 50 days ma and its up 3 days in a row. Sell the open on the 4th day and buy the close.
I have a ton of these types of systems. Some better than others. My goal was to get some feedback on how to improve it. With one suggestion Nitro made it much much better.
Maybe you could take a look and provide some feedback as well. Thx for taking a look. Jim

jim c · Aug 3, 2009

Quote from Technician:

Can you explain how this works? I couldn't figure it out.
More...

I dont have fair value data so I wasnt able to test this completly. I just added the parm that vix had to be greater than 20 to place a trade. I was suprised at the results. This is exactly the type of advice I was looking for (thx Nitro). Thx for taking a look . JIm

Log in or Sign up

System testing questions.

jim c

TrueStory

Technician

alpri16

jim c

jim c