Re: [ojAlgo-user] QuadraticSolver issue

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

Here's my code. testP20110614 is your *problem*

http://ojalgo.cvs.sourceforge.net/viewvc/ojalgo/TestProj/src/org/ojalgo/finance/portfolio/ReportedProblems.java?view=markup

/Anders


On 14 jun 2011, at 23.33, Simon Greaves wrote:

> Yes, I have to finish the lower bounds, that part I'm not worried about.
> But I don't understand, I must be doing something different from how you did it, because I get:
> 
> X class org.ojalgo.matrix.PrimitiveMatrix
> {{1.000000000000013},
> {3.2333333333333143},
> {-3.233333333333325}}
>  
> This is how I get X:
> 
> Result rslt = qdrtc_slvr.solve();
> BasicMatrix rslt_mtrx = rslt.getSolution();
> PhysicalStore<Double> result = rslt_mtrx.toPrimitiveStore();
> System.out.println("\nX " + rslt_mtrx.toString());
> 
> I this correct?
> Thanks again for all your help by the way!
> 
> SG
> 
> From: Anders Peterson <an...@op...>
> To: oja...@li...
> Sent: Tue, June 14, 2011 4:34:59 PM
> Subject: Re: [ojAlgo-user] QuadraticSolver issue
> 
> Here's what I get:
> 
> [AE] = class org.ojalgo.matrix.PrimitiveMatrix
> {{1.0,    1.0,    1.0}}
> 
> [BE] = class org.ojalgo.matrix.PrimitiveMatrix
> {{1.0}}
> 
> [Q] = class org.ojalgo.matrix.PrimitiveMatrix
> {{4.0,    1.0,    2.0},
> {1.0,    9.0,    1.0},
> {2.0,    1.0,    16.0}}
> 
> [C] = class org.ojalgo.matrix.PrimitiveMatrix
> {{10.0},
> {15.0},
> {18.0}}
> 
> [AI] = class org.ojalgo.matrix.PrimitiveMatrix
> {{1.0,    0.0,    0.0},
> {0.0,    1.0,    0.0},
> {0.0,    0.0,    1.0}}
> 
> [BI] = class org.ojalgo.matrix.PrimitiveMatrix
> {{1.0},
> {1.0},
> {1.0}}
> 
> [X] = class org.ojalgo.matrix.PrimitiveMatrix
> {{-0.17441860465116266},
> {0.6279069767441859},
> {0.5465116279069767}}
> 
> [LE] = class org.ojalgo.matrix.PrimitiveMatrix
> {{8.976744186046512}}
> 
> [LI] = class org.ojalgo.matrix.PrimitiveMatrix
> {{0.0},
> {0.0},
> {0.0}}
> 
> [SE] = class org.ojalgo.matrix.PrimitiveMatrix
> {{0.0}}
> 
> [SI] = class org.ojalgo.matrix.PrimitiveMatrix
> {{1.1744186046511627},
> {0.37209302325581406},
> {0.4534883720930233}}
> 
> I do not get any weights larger than 1.0, but I do get one weight smaller than 0.0. There's no constraint to restrict that! Add constraints for that and you should get the solutions you expect.
> 
> Here's how I would find the efficient frontier:
> 
>         final BasicMatrix tmpCovars = ...;
>         final BasicMatrix tmpReturs = ...;
> 
>         final MarketEquilibrium tmpME = new MarketEquilibrium(tmpCovars);
> 
>         final MarkowitzModel tmpMarkowitz = new MarkowitzModel(tmpME, tmpReturs);
> 
>         for (int i = 1; i < 10; i++) { // Some different risk aversion factors
> 
>             tmpMarkowitz.setRiskAversion(new BigDecimal(i));
> 
>             final List<BigDecimal> tmpWeights = tmpMarkowitz.getWeights();
>         }
> 
> ------------------------------------------------------------------------------
> EditLive Enterprise is the world's most technically advanced content
> authoring tool. Experience the power of Track Changes, Inline Image
> Editing and ensure content is compliant with Accessibility Checking.
> http://p.sf.net/sfu/ephox-dev2dev_______________________________________________
> ojAlgo-user mailing list
> ojA...@li...
> https://lists.sourceforge.net/lists/listinfo/ojalgo-user