- ago


Column 2 is the APR and Column 3 is the Net Profit. Larger profit but lower APR?

I think there is a bug somewhere in the calculation.

Vince
0
173
10 Replies

Reply

Bookmark

Sort
GlitchA
 ( 11.34% )
- ago
#1
Thank you for the report.
0
GlitchA
 ( 11.34% )
- ago
#2
Although, it could be explained by a difference in StartIndex.

Strategies that evolve with larger period indicators will have a larger StartIndex, so the range of data used to calculate the annualized return is less.

If you double click on a Strategy you can run it in a Strategy window to validate the APR.
0
- ago
#4
I don't think that is the case


The large Profit cases are using Gaussian MAs of ~50bars and the smaller Profits are using 50/200 SMA XOvers.

Vince
0
GlitchA
 ( 11.34% )
- ago
#5
Have you double clicked one of the suspect Strategies to see that the metrics show in the Strategy window? So far I haven't been able to duplicate this.
0
- ago
#6
I noticed something that appears to be common to all of the entries. They are all built on the McGinley Dynamic.

Could that be an issue?

And yes, the metrics are (more or less) the same in a backtest compared to the Evolver.

Vince
0
- ago
#7
And I just noticed another bit of strangeness. The Profits and APR are properly "aligned" at the start of an Evolver run but then start to become "disconnected" after running for many Generations.

Vince
0
GlitchA
 ( 11.34% )
- ago
#8
Thanks for the additional info, let’s see if I can figure out what’s going on here! Do any other metrics have anomalies?
0
- ago
#9
There does not appear to be any other discrepancies.

Vince
0
- ago
#10
QUOTE:
Glitch: Strategies that evolve with larger period indicators will have a larger StartIndex, so the range of data used to calculate the annualized return is less.


After spending a LOT of time chasing the problem, I have come to conclusion that you are correct. Sorry for the wild goose chase.

Vince
0

Reply

Bookmark

Sort