USERS. How do you resolve "market" order backtesting if you dont have bid/ask data?

[maxima]

As long as I understand - during backtests Ninja "execute" market orders at Last price.

I dont have bid/ask data in my files. So I suppose is that most of times NT strategy run on such data will be wrong at least at 1 tick (if Last was Best Bid for instance and NT strategy analiser is trying to Exit Long using Last but in reality it would be at least one tick worse)...

Is that correct understanding?

How do you normally backtest system to give it realistic market order execution? Do you set Slippage to 1 ? How good it correlate with usual CME markets (ES,NQ,ZB,6E) ?

[NinjaTrader_Austin]

maxima, the default behavior for the backtester is to fill the order at the price of the open of the next bar. Increasing slippage is usually a good way to get more realistic results.

[maxima]

Thanks Austin, I know that, so I always use second time-frame - 1 tick to "execute" orders. But still at next tick NT doesnt know what is the best bid/ask....

I thought there must be quite a lot of people involved in backtesting of various systems, and the question of "realism" of it must be the first thing to resolve....

I hoped that someone might be willing to share a little insight perhaps.

[NinjaTrader_Bertrand]

As Austin mentioned, increasing slippage and or commissions is a good way to gauge if you have a solid edge under adverse conditions - NinjaTrader 7 allows for storing and accessing of historical bid / ask data so have a better 'grip' on this then...

http://www.ninjatrader.com/webnew/NT7/NinjaTrader7.html