Hi there,
After optimizing an EMA based strategy, I have noticed that some of the numbers proposed by the optimization are quite illogical. It seems to swop the EMA numbers around by making the Slow EMA number lower than the fast EMA. As a result the metasentimentor seems to take long positions when the market is dropping and short positions when the market is rising. Is this correct and can I trust the optimization when it does this even though it shows profitability of over 80% on many sets of optimized numbers?
Optimizing EMA for automated trading
Optimizing EMA for automated trading
You do not have the required permissions to view the files attached to this post.
- WHS Support
- Posts: 2433
- Joined: 14 Feb 2013 10:27
Re: Optimizing EMA for automated trading
Hello Crighton,
The optimization looks for the mathematical best solution. It does not consider if the numbers are logical.
So it might be the case that you describe that actually selling on a "normal" buy signal might lead to a better result.
You can use the highest and lowest settings for the indicator in the Designer Bar to "guide" the optimization to a more realistic result. For example you could say: Slow EMA, between 25 and 100, Fast EMA between 5 and 20. Then it will look for a result within these parameters.
Kind regards
Dominic
WHS
The optimization looks for the mathematical best solution. It does not consider if the numbers are logical.
So it might be the case that you describe that actually selling on a "normal" buy signal might lead to a better result.
You can use the highest and lowest settings for the indicator in the Designer Bar to "guide" the optimization to a more realistic result. For example you could say: Slow EMA, between 25 and 100, Fast EMA between 5 and 20. Then it will look for a result within these parameters.
Kind regards
Dominic
WHS
Re: Optimizing EMA for automated trading
Thanks, that makes a lot of sense.