Can't Believe Mean Reversion Backtest Results

tip to OP......can the ALL CAPS...

now that we have that out of the way.....

your backtest time period is probably not long enough....try a randomly different time period like maybe 2008, 2009, 2007....pick one or two and see what you come up with....

Don't get too overly excited because, most likely, all you have is one gigantic curve-fit..

If you insist on moving forward, then take the trades on sim for 3-4 months to see if it blows up immediately in your face...then start small....although with the number of trades that your system takes...starting small won't prevent you from incurring huge losses if/when the system blows up unless maybe you're performing realtime statistical analyses on the drawdown sizes...
 
lol

NT cannot test on BID&ASK, it tests on just BID.
Average holding period 1min....

Tells me NT fools you. The fills will not work in reality.
 
THANK YOU ALL FOR REPLIES EXTREMELY HELPFUL AND I SHALL TAKE HEED TO THE COMMENTS SPECIFICALLY RUNNING A LIVE SIMULATION FOR A FEW MONTHS UPPING MY SLIPPAGE AND TRYING TO WORK ON LARGEST DRAW-DOWN ANY SLIPPAGE SETTING SUGGESTIONS FOR A STRATEGY LIKE THIS PLUS ANY QUALITY READING ON STRATEGY DEVELOPMENT AND REFINEMENT ?
 
Quote from omega_350:

THANK YOU ALL FOR REPLIES EXTREMELY HELPFUL AND I SHALL TAKE HEED TO THE COMMENTS SPECIFICALLY RUNNING A LIVE SIMULATION FOR A FEW MONTHS UPPING MY SLIPPAGE AND TRYING TO WORK ON LARGEST DRAW-DOWN ANY SLIPPAGE SETTING SUGGESTIONS FOR A STRATEGY LIKE THIS PLUS ANY QUALITY READING ON STRATEGY DEVELOPMENT AND REFINEMENT ?
More...
YOU MIGHT ALSO WANT TO IMPLEMENT A NOISE REDUCTION FILTER. i DON'T REALLY UNDERSTAND WHAT YOU ARE DOING, BUT IT SOUNDS VERY LOUD TO ME
 
Quote from omega_350:

I FINALLY STARTED TRADING STRICTLY WITH ALGORITHMS FOR MY ENTRIES AND I DO THE REST AND I HAVE 14 DAYS WITH OUT A SINGLE LOSING DAY AND VIRTUALLY NO STRESS ANYWAY I HAVE DESIGNED A MEAN REVERSION STRATEGY FOR THE ES AND ZB AND I CANT BELIEVE MY RESULTS ON EITHER ONE ESPECIALLY THE ES PLEASE SOMEONE REVIEW FOR COMMON ERRORS AND SUGGEST ANYTHING I MIGHT BE MISSING
More...

Average trade is $6.80. Stop markets will cause slippage, so merely adding 1 tick per side will make this a losing strategy.

I see you have a $5 commission per trade, and I don't know which broker you expect to get that from, but that seems to be on the low side.

If I missed limit order entries and stop market exits the stop market exits or stop market entries guarantee 1 tick of slippage to happen fairly often.

Your broker will love you, but you will not make money with this due to slippage. Even if you're using limits consider the connectivity of your machine to the internet as another possible source of slippage that could exceed many times more than a single tick of slippage.
 
Speaking of fills and slippage, I wonder how the testing went in August with the high volatility.

And then the obvious question is, how would it do in an extremely fast market such as the fall of 2008. For some, those months made their career. For others, it was the end of the game.
 
here is a benchmark... (btw keep working on your system)

so lets say

systems nets 100k
systems gross 200k
systems loss 100k

pf 2.0 this will survive (maybe)
 
You must log in or register to reply here.
Back
Top