Comment by Peter jcp on February 5, 2012 at 10:33am

Hi Max - I like the quote below - its true as long as the 48% of the trading losses are kept lower than the 52% trading wins -

 

Profitable market dynamic identified. "If your trading model gives you an edge of even just 2% over 50/50, that is enough to make a substantial profit if you trade frequently enough," notes Rezania."Using MathWorks tools, we developed and backtested some strategies that provide that statistical edge."

Also very worried if they have only been making 14% net per annum on currencies - something is wrong there - HFT should be making about 14% per week (lol)- but no professional body would dare say that. - regards Peter

Comment by Max Bongo on February 5, 2012 at 11:32am

Yes they may have opted to write some ambiguous text for their appraisal on the software' developer's website? I was also bemused by, "2% over 50/50," they must have a very minimal loss on the losses to make any advantage of this.

Comment by 2ndSkiesForex on February 5, 2012 at 2:55pm

@Peter

You have to understand the nature of CalPERS as an investment vehicle.  They manage pension funds, so cannot be going for 14% a month because it carries that much downside risk.  They have objectives and an umbrella of rules for which they have to abide by, hence why they cannot go for those levels of profit because they carry similar levels of risk.

I have actually talked to a couple of people at CalPERS as they are more interested in the 6-25% range per year with very low risk, and very low drawdowns.  But, they have loads of cash (billions) so doing those kinds of returns would yield a handsome profit.

My programmer uses MATLAB and that thing is badass.  The stuff we are able to do with it is amazing, especially when doing dynamic economic modeling.

Chris Capre
2ndSkiesForex.com
Twitter; 2ndSkiesForex 

Comment by Peter jcp on February 5, 2012 at 5:49pm

Hi Chris - I appreciate commercial bodies are bound by more rules and regulations and if they are managing billions, a very low risk element must be paramount. However new boundaries are be broken with this technology and similar methods out of the larger commercial restraints could rewrite the ROI "guide / acceptance" books. I just now wonder how this will be controlled or regulated during the next few years - might it actually become banned ??

Comment by 2ndSkiesForex on February 5, 2012 at 6:18pm

Yes, that is true, boundaries are being broken, but the bottom line is many of the HFT algo's wouldn't comply with the risk parameters funds like CalPERS can work with.  You have to understand when algos go wrong, they go very wrong.  Why was the 2008 crash so bad?  A large part of that was hedge funds which were running algos, many HFT's, that were using algo's which could not deal with the volatility or 6+sigma nature of the market movements.  So they kept getting into positions that were getting crushed, they kept buying when the market was collapsing for months.  No fund like CalPERS can take on risk like that.  They have strict rules they have to comply with and that is what they base their decisions on...not just ROI.

Kind Regards,
Chris Capre 

Comment