When I heard NOK/SEK had dropped lately I realized it was the same last year, and the year before that, so I made a simple backtest and thought the results were nice compared to the simplicity. It might not be much for automation as much as a concept to trade manually, but still sharing here if anyone has any input on how it could be improved and automated. Or if there are other pairs that perhaps has a stronger effect. I wanted to add stop loss etc but I didn’t manage adding a max trades per month properly. Also not sure what the spread is so I set it at 6
shorttime = month = 11
longtime = month=1
exitlongtime = month>=5
exitshorttime = month=1
//higherlow = low > highest[x](low[1])
//lowerhigh = high < lowest[y](high[1])
// Conditions to enter long positions
IF NOT LongOnMarket AND longtime THEN
BUY 8 CONTRACTS AT MARKET
ENDIF
// Conditions to exit long positions
If LongOnMarket AND exitlongtime THEN
SELL AT MARKET
ENDIF
// Conditions to enter short positions
IF NOT ShortOnMarket AND shorttime THEN
SELLSHORT 8 CONTRACTS AT MARKET
ENDIF
// Conditions to exit short positions
IF ShortOnMarket AND exitshorttime THEN
EXITSHORT AT MARKET
ENDIF
It seems a good performance, but keep in mind that you used too small an initial capital. If you want to double that performance you just have to write SEK 3000, instead of 6000!
A better initial capital would be 3 to 4 times the DD, that is roughly SEK 40-50000, thus reducing gaining percentage from 850% to about 140% in 13.5 years, which is 10%/year. Moreover did you enable tick-by-tick mode? I’m afraid not, because bactkesting wouldn’t have started back from 2006, I guess. Try enabling it and using a larger starting capital.