Would You so kind to show how can be automated the process of genetic optimization by DataSets and assign each ticker in slected DataSet different PreferedValues? Maybe You have some example codes? Thank You in advance!

The process can not be automated, because a) automating optimizers (whatever it may mean) is not supported and b) we do not expose any APIs for "driving" the GA optimizer. However, maybe you could clarify what do you mean by saying "assign each ticker in slected DataSet different PreferedValues?" How exactly does it differ from assigning PVs based on the highest/lowest average metric value for all symbols?

Thank you for fast answer, Eugene!
I specify question in other way - How can I use a WealthLab libraries in my own application written in C# or VB.NET for optimization calculations? I have read the Advanced Programming Topics - "Creating Optimizers in Wealth-Lab", but can't find none typical example codes. Maybe you have something for faster insight in this task?

QUOTE:
How exactly does it differ from assigning PVs based on the highest/lowest average metric value for all symbols?

I use combination of several parameters (Net Profit, Trades, Sharpe, Ulсer index) to determine what values ​​to set as PreferedValue for each Ticker. Since WLD 6 does not offer this option, I have to build additional software to make this computation.

QUOTE:
How can I use a WealthLab libraries in my own application written in C# or VB.NET for optimization calculations?

A rule of thumb is that if you can't find something covered in the documentation, which is the QuickRef and Wealth-Lab Development Guide (a.k.a. the API manuals), then it means that this something is not supported. Specifically, using Wealth-Lab and/or MS123 libraries in 3rd party applications is not supported. You won't find any examples of it, sorry.

Hi,

I am trying to code an algorithm where by:
at bar=0: the code optimizes the strategy parameters of the indicator at t0;
bar 0 --> 500: the code trades this indicator at given (optimized) setup, for next 500 bars;
at bar 500: THE CODE OPTIMIZES THIS INDICATOR,, AGAIN, USING LAST 500 bars' data;
bar 500 --> 1000: the code continues to trade this indicator, now with newer set of parameters;
bar 1000: optimization again;
bar 1000 > 1500: trade given new set of strategy parameters;
...
...
until the last bar.


**

Is there any example code, or a wiki link, or another customer ticket, that I can refer to in order to code this type of trading algorithm?

Given that there is walk-forward-optimization property of the platform, I believe there are classes, methods, properties and all sort of technical base to code such 'optimize-on-the-fly' kind of algorithms.

**

And as usual, thanks, again, for your consideration.

Aykut

Hi Aykut,

You won't find any examples of leveraging any internal classes to perform on-the-fly/programmatic optimizations. However, what you described is a perfect case for WFO optimization. Then why are you asking about doing it the hard way when it's already available?

Thank you for the reply, Eugene.

I will go over again the WFO and its capabilities; i think i did not understand fully yet about what this extension is capable of.

Best regards,