Hi,
I've just created a system in Multicharts with IOG turned on, tried to use it in Portfolio then learnt I need to have it turned off to work in Portfolio.
With IOG turned on I use "close" to represent the currentprice e.g. if close > high[1] then do something.
However, when you have IOG turned of, "close" now represents the close value of the currentbar which changes the outcome of my code.
1. What are some tips to ensure accurate, close to real-life trade behaviour with IOG turned off?
2. When will Portfolio support IOG and barmagnifier?
Regards
Champski
How to get accurate, close to real-life results with intrabarordergeneration = false
- TJ
- Posts: 7740
- Joined: 29 Aug 2006
- Location: Global Citizen
- Has thanked: 1033 times
- Been thanked: 2221 times
Re: How to get accurate, close to real-life results with intrabarordergeneration = false
1: use a finer resolution.
-
- Posts: 275
- Joined: 22 Apr 2014
- Has thanked: 70 times
- Been thanked: 72 times
Re: How to get accurate, close to real-life results with intrabarordergeneration = false
In theory TJ is right. In practice it is very difficult. I have used PT for my primary trading for a couple of years now with IQFeed and I am trading all 100 of the S&P100 (not all at the same time). Using minute data is fine for trading but I find it terribly inaccurate for back testing. Data loading is also very slow. Dropping down to second data only works with a minimal amount of history. You might get it to work with just a few instruments but if you are trying to trade a whole index of stocks, it is impractical.
I think you can get a rough estimate that your strategy won't kill you but your only real option is to forward test and log all your trades externally. The difference between real time and back test can be huge. Things happen in real time that never get picked up by the back tester.
I like PT for executing trades and running my strategy. It does this flawlessly but I think you have to be very careful how you use it as a testing tool.
Regards,
Alex
I think you can get a rough estimate that your strategy won't kill you but your only real option is to forward test and log all your trades externally. The difference between real time and back test can be huge. Things happen in real time that never get picked up by the back tester.
I like PT for executing trades and running my strategy. It does this flawlessly but I think you have to be very careful how you use it as a testing tool.
Regards,
Alex
Re: How to get accurate, close to real-life results with intrabarordergeneration = false
Thanks hughesfleming.In theory TJ is right. In practice it is very difficult. I have used PT for my primary trading for a couple of years now with IQFeed and I am trading all 100 of the S&P100 (not all at the same time). Using minute data is fine for trading but I find it terribly inaccurate for back testing. Data loading is also very slow. Dropping down to second data only works with a minimal amount of history. You might get it to work with just a few instruments but if you are trying to trade a whole index of stocks, it is impractical.
I think you can get a rough estimate that your strategy won't kill you but your only real option is to forward test and log all your trades externally. The difference between real time and back test can be huge. Things happen in real time that never get picked up by the back tester.
I like PT for executing trades and running my strategy. It does this flawlessly but I think you have to be very careful how you use it as a testing tool.
Regards,
Alex