Running Multicharts signal with two sets of parameters
I have a signal that produces nice results with a certain set of parameters, call it set A.
The same signal produces good results with another set of parameters, call it set B.
Running both signals in parallel should, according to my backtesting, improve returns to some extent while decreasing drawdown, thereby improving the Return On Account significantly.
However when I tried to do that I get double exit orders. For example - when the signal with set A enters a position, then both signals place stop and limit exit orders. So I have a position of 1 lot with two stop orders of 1 lot each and two limit orders of 1 lot each.
Is there a simple way to run two signals on the same instrument and have them completely independent from each other. Perhaps in Portfolio trader?
it sounds like you placed both signals on the same chart and your code is causing this behavior.
I don't think it will work without modifications in Portfolio Trader - same thing would be two separate charts, you'd probably end up with a different result than what you'd get combined on one chart. Therefore you will probably need to adjust your code to manage only the positions you want it to, tie exits to specific entries etc..
The following user says Thank You to ABCTG for this post:
If they're going to be on different charts - will they run completely independent of each other? mind you, I am using marketposition(0) to determine if I have an order and if it's long or short, and send orders accordingly.
So I guess my question is, does marketposition and similar orders return the status on the broker side, or according to the code on the specific chart.
My purpose is to run them completely independent of each other. As if they were running on different accounts.
Last edited by TraderWalrus; September 1st, 2016 at 06:47 AM.
This depends on your code. If you don't use anything that has the ability to exchange data between charts or use reserved words that use the market position at the broker for example, then the results should be independent from each other. But it's likely that the results will be different than from your backtest, too - simply because you are looking at separate systems (as in the combination of signals one might block the other, which wouldn't be the case when you run it on separate charts).
If your backtests have been done by a combination of the signal on one chart, I'd invest the time to track down what causes the situation you encounter live. If you split the system now or switch to portfolio trader you create a new situtation and it might not have any relevance to your backtest anymore.