FYI, I put more thought into this and ended up writing Excel VBA code to create
Amibroker code that is like the code I show in my previous post. It parses the
start and end dates as well as the parameter names and values off of the walk
forward report.
However, after all this work, I realized I can not easily substitute changing
parameter values because I am using AB's built-in array functions for certain
calculations, and these functions accept SINGLE values for lengths/inputs
whereas I would need them to accept ARRAYS for the changing lengths.
So now I am recoding the few functions I need into variable-length versions
that match the result of the built-in array functions. After doing so, I should
be able to generate an accurate out-of-sample backtest Report for the entire
out-of-sample concatenated series.
Paul
--- In amibroker@yahoogroups.com, "notanaiqgenius" <notanaiqgen...@...> wrote:
>
> Hi,
>
> Does anyone know a way to generate a full backtest report specifically for
> the ENTIRE out-of-sample concatenated backtest?
>
> I understand and know how to access the composite tickers such as ~~~ISEQUITY
> and ~~~OSEQUITY, but I don't see a way I could then generate ONE full
> backtest report for entire the concatenated out-of-sample pieces.
>
> The only way I can think it might be done is to manually write code at the
> top of my AFL using dates and variable values from the walk forward report
> like:
>
> dn = DateNum();
> for(i=0;i<barcount;i++)
> {
> if(dn[i]>=1030101 & dn[i]<1030701)
> {
> //assign optimized inputs to values chosen for this OS segment
> optedVar1[i] = 5;
> optedVar2[i] = 10;
> }
> if(dn[i]>=1030701 & dn[i]<1040101)
> {
> //assign optimized inputs to values chosen for this OS segment
> optedVar1[i] = 7;
> optedVar2[i] = 13;
> }
> if(dn[i]>=1040101 & dn[i]<1040701)
> {
> //assign optimized inputs to values chosen for this OS segment
> optedVar1[i] = 11;
> optedVar2[i] = 12;
> }
> //ETC. rolling forward through each out-of-sample segment
> }
>
> Anyone have a better way to do this?
>
> Thank you- I appreciate your time reading this.
>
> Paul
>
