Re: [ojAlgo-user] Markowitz optimisation

[Anders P. <an...@op...>]

On 17 jan 2008, at 01.38, Arthur McGibbon wrote:

> Hi Anders,
>
> thanks for your reply and the link.
>
> I am indeed trying to maximise sharpe ratio (at zero risk free  
> rate) using your library.
>
> I can run the MarkowitzModel.calculatePortfolioWeights method fine  
> as is but I'm trying to apply weighting constraints to the assets  
> (limits of 5% -> 55% with 10 assets).  I looked in the  
> MarkowitzModel code and saw that the limitation on shorting was  
> applied by using the setLowerLimit(ZERO) on the variables and  
> assumed that I could use the same method in conjunction with the  
> setUpperLimit (passing values of 0.05 and 0.55) however this  
> results in NullPointerException on certain combinations of returns.

Can you give me a code example that generates a NullPointerException?

> I mentioned Risk Aversion Factor before because I assumed my error  
> was there as some RAFs generated exceptions and some didn't but it  
> looks as though my assumption on how to apply weighting limits was  
> incorrect so could you tell me whether it's possible to apply min  
> and max weighting limits to the assets using ojAlgo?

I don't think your assumption is wrong, and I'd like to know about  
those exceptions. Give me code snippets that I can run.

/Anders


> -----Original Message-----
>> From: oja...@li...
> [mailto:oja...@li...]On Behalf Of Anders
> Peterson
> Sent: 16 January 2008 22:41
> To: oja...@li...
> Subject: Re: [ojAlgo-user] Markowitz optimisation
>
>
> On 16 jan 2008, at 15.36, Arthur McGibbon wrote:
>> Hi,
>>
>> I've been using the MarkowitzModel code to optimise the weighting
>> across assets within a portfolio but what I ideally want is the
>> weights for the maximum return/risk ratio possible with the
>> constraints passed (i.e. the maximum slope on the efficient
>> frontier going through the origin).
>>
> http://en.wikipedia.org/wiki/
> Modern_portfolio_theory#The_market_portfolio
>> Is it possible to use the QuadraticModel to solve for this maximum
>> given the expected returns and covariance matrices?
>> If so - could you give me some pointers on how to do this?
>> Also - would this lead to not using a Risk Aversion Factor as the
>> maximum return/risk ratio should lead to a single point on the line
>> or have I got the wrong idea about what the Risk Aversion Factor is?
>>
> I think you're the one who should know which (mathematical) problem
> you want to solve.
>
> When if you have more specific questions related to ojAlgo source
> code I'll be glad to try and help.
>
> /Anders
>
> ---------------------------------------------------------------------- 
> ---
> This SF.net email is sponsored by: Microsoft
> Defy all challenges. Microsoft(R) Visual Studio 2008.
> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
> _______________________________________________
> ojAlgo-user mailing list
> ojA...@li...
> https://lists.sourceforge.net/lists/listinfo/ojalgo-user
>
>
> ********************************************************************
> This message and any attachment are confidential.  If you are not  
> the intended recipient please contact the sender, delete this  
> message and any attachment from your system and do not disclose,  
> copy or distribute the contents to any other person.  Beach Horizon  
> LLP (Beach Horizon) is not responsible for any information  
> contained in this email.  Beach Horizon reserves the right to  
> monitor all email messages passing through its network.
> ********************************************************************
>
>
> ---------------------------------------------------------------------- 
> ---
> This SF.net email is sponsored by: Microsoft
> Defy all challenges. Microsoft(R) Visual Studio 2008.
> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
> _______________________________________________
> ojAlgo-user mailing list
> ojA...@li...
> https://lists.sourceforge.net/lists/listinfo/ojalgo-user
>
>