Is there any way to trade against multiple pairs within a given backtest? I'm admittedly uninformed on this, but I've noticed that when I run a backtest in MT4 that I always have to run it against a specific pair.
I'm asking because it can be very difficult to determine a realistic expectation for drawdown unless you can test multiple pairs concurrently. For example, take the following set of theoretical trades:
Buy USDJPY, closed @ +100
Buy GBPUSD, closed @ -50
Sell USDJPY, closed @ +150
Sell GBPUSD, closed @ -50
Buy USDJPY, closed @ -50
Buy GBPUSD, closed @ +50
Sell USDJPY, closed @ +100
Sell GBPUSD, closed @ -50
Buy USDJPY, closed @ +200
Buy GBPUSD, closed @ -100
In this example, GBPUSD suffered a drawdown (-200) but the system performed very well (+300) when balanced against the trades that were taken on USDJPY (+500). If I were just looking at GBPUSD, I might think that the system is ineffective. Similarly, the equation could be reversed during other periods of time, with USDJPY's losses being offset by wins on GBPUSD. This type of inter-pair relationship is a real pain to discern if I have to perform all backtests against single pairs and then attempt to manually intersperse the trades on a chronological timeline.
My comparison would be when doing backtests against stock market data. Most stock market backtesting programs are not designed to be run against a single stock. Rather, you tell the program what criteria to use when picking trades from the entire market, and then you view the aggregate result of all those trades against many different stocks.
I'm asking because it can be very difficult to determine a realistic expectation for drawdown unless you can test multiple pairs concurrently. For example, take the following set of theoretical trades:
Buy USDJPY, closed @ +100
Buy GBPUSD, closed @ -50
Sell USDJPY, closed @ +150
Sell GBPUSD, closed @ -50
Buy USDJPY, closed @ -50
Buy GBPUSD, closed @ +50
Sell USDJPY, closed @ +100
Sell GBPUSD, closed @ -50
Buy USDJPY, closed @ +200
Buy GBPUSD, closed @ -100
In this example, GBPUSD suffered a drawdown (-200) but the system performed very well (+300) when balanced against the trades that were taken on USDJPY (+500). If I were just looking at GBPUSD, I might think that the system is ineffective. Similarly, the equation could be reversed during other periods of time, with USDJPY's losses being offset by wins on GBPUSD. This type of inter-pair relationship is a real pain to discern if I have to perform all backtests against single pairs and then attempt to manually intersperse the trades on a chronological timeline.
My comparison would be when doing backtests against stock market data. Most stock market backtesting programs are not designed to be run against a single stock. Rather, you tell the program what criteria to use when picking trades from the entire market, and then you view the aggregate result of all those trades against many different stocks.