Account Performance Backtesting

[motownlucas]

Is there any way to look at simulated account performance of a back-tested strategy?

I'd like to backtest a strategy, or multiple strategies, on a basket of instruments, say the S&P 500, and then view the hypothetical account performance.

As it stands, if I backtest on those 500 instruments the results assume that I take every trade regardless of available account equity at time of entry. Also, Monte Carlo is only available one instrument at a time.

[NinjaTrader_Brett]

Hello,

Thanks for the forum post.

You can run the tests on each instrument in a basket as you mentioned all at once, However there is no cross communication from all the strategies. Each strategy ir run as if it was its own as if it was the only strategy running.

Let me know if I can be of further assistance.

[motownlucas]

Thanks Brett.

Ok, so forget multiple strategies...

I can test multiple instruments at once, but can I do it as an account simulation where a new strategy position will only be entered if there is available account equity not already committed to a position on another instrument? And then the trades/results across all instruments can be analyzed together, run through Monte Carlo, etc.?

[NinjaTrader_Brett]

Understood.. The backtester does not take into effect account equity and will do the trades as you have coded them.

If you needed this functionality in your backtest you would need to code this in by setting a value at the start of your strategy as your account equity and then subtracting or adding from it as trades complete.

An example that can be modified to fit your needs would be here:

http://www.ninjatrader.com/support/forum/showthread.php?t=19289

Let me know if I can be of further assistance.

[motownlucas]

Each instrument or chart runs it's own separate instance of the strategy, so how would I code something in the strategy that looks at trades and account equity across a basket of instruments in a strategy analyzer backtest?

[motownlucas]

In real time the strategy could look at current available funds and reject new positions accordingly. Far as I can tell the backtest runs each instrument sequentially in a vacuum as opposed to simultaneously and chronologically as if it were playing out in real time. So I don't see any way to reject a new position in one instrument due to being fully subscribed in other instruments during a basket backtest. Is that right?

[NinjaTrader_Brett]

Hello,

That is correct. In realtime it would work correctly however with backtest it is fundamentally left out as to not cause issues. For example if we allowed this then we would run into issues where based on user settings for their simulator account would run into completely different strategy results and trades based on this one setting which is not desired since most traders want the strategy to run the same way every time. Whether there running it today or 2 weeks from now backtesting.

Let me know if I can be of further assistance.