Re: [ojAlgo-user] Potential bug in org.ojalgo.finance.FinanceUtils.makeCovarianceMatrix() method
Mathematics, linear algebra and optimisation
Brought to you by:
apete
From: Anders P. <an...@op...> - 2013-06-28 07:08:48
|
Have a look at this application: http://t-cloud.blaadin.com/ (You need to log in using your e-mail address.) Under "Define Market" and "Setup Profiles" you'll see what the risk aversion is for the market portfolio and the various risk profiles. On the "Assign Confidence" page you'll se all efficient portfolio weights. I used a 0.1 to 100 range on the risk aversion to create that chart. /Anders On 27 jun 2013, at 23:51, Anders Peterson <an...@op...> wrote: > The risk aversion factor is up to you... ;-) > > 2 and 4 are reasonable values, but as the entire range it's a bit narrow. 1 to 10 will cover a bit more, and 0.1 to 100 will usually cover just about everything. Try different values and see what results you get. Also try the calibrate(...) methods of FixedReturnsPortfolio and FixedWeightsPortfolio. > > /Anders > > > On 27 jun 2013, at 19:09, Yang Huang <yan...@gm...> wrote: > >> Hi Anders, >> >> Thanks for the quick fix. It works fine now. However, I realized that I have another simple question regarding the Risk Aversion Factor of the several portfolio classes, e.g. MarkowitzModel and MarketEquilibrium. What is the range for this factor? In some cases, a practical range for the factor can be between 2 to 4. I couldn't find a clue how this factor can be defined. Could you please also advise that? Thanks. >> >> -- >> Yang >> ------------------------------------------------------------------------------ >> This SF.net email is sponsored by Windows: >> >> Build for Windows Store. >> >> http://p.sf.net/sfu/windows-dev2dev_______________________________________________ >> ojAlgo-user mailing list >> ojA...@li... >> https://lists.sourceforge.net/lists/listinfo/ojalgo-user > |