Re: [ojAlgo-user] Markowitz Model Problem
Mathematics, linear algebra and optimisation
Brought to you by:
apete
Menu
▾
▴
Re: [ojAlgo-user] Markowitz Model Problem
|
From: Anders P. <an...@op...> - 2009-01-15 13:09:57
|
Your expected excess returns matrix looks like this:
class org.ojalgo.matrix.BigMatrix
{{-0.010638291263564232},
{-0.013500370827906071},
{-0.011390037735101773},
{-0.010385042339767682},
{-0.0003812208389845893},
{-0.002315505853720011},
{0}}
Is that correct?
Earlier you wrote "And i found that if i change the nagetive
ExceptedReturn to positive, the result will be fine."
/Anders
On 15 jan 2009, at 11.10, Ruan Weichao wrote:
> Hi，i am using v25 and here's my code:
>
> /
> **********************************************************************
> **********/
>
> import java.math.BigDecimal;
> import java.util.List;
>
> import org.ojalgo.finance.portfolio.MarketEquilibrium;
> import org.ojalgo.finance.portfolio.MarkowitzModel;
> import org.ojalgo.matrix.*;
> import org.ojalgo.matrix.store.BigDenseStore;
> import org.ojalgo.matrix.store.MatrixStore;
>
>
> public class TestMarkowitz {
>
> public double getCovarance(double[] value1, double[] value2){
>
> int n = value1.length;
> double averOne = 0;
> double averTwo = 0;
> for(int i=0;i<n;i++)
> {
> averOne+=value1[i];
> averTwo+=value2[i];
> }
> averOne /= n;
> averTwo /= n;
>
> double sum = 0;
> for(int i=0;i<n;i++)
> sum+=(value1[i]-averOne)*(value2[i]-averTwo);
>
> return sum / (n-1);
> }
>
> public BasicMatrix getACovariances(double[][] returns){
>
> int row = returns.length;
> int col = returns[0].length;
> MatrixStore<BigDecimal> a = BigDenseStore.FACTORY..makeZero(row,
> col);
> BasicMatrix covariances = new BigMatrix(a);
>
> for(int i=1;i<=row;i++){
> for(int j=i;j<=col;j++){
> double tmp = this.getCovarance(returns[i-1], returns[j-1]);
> covariances = covariances.set(i-1, j-1,tmp);
> covariances = covariances.set(j-1, i-1,tmp);
> }
> }
> return covariances;
> }
>
> public BasicMatrix getAnExpectedExcessReturns(double[][] returns){
>
> int row = returns.length;
> int col = returns[0].length;
>
> MatrixStore<BigDecimal> a = BigDenseStore.FACTORY..makeZero(row, 1);
> BasicMatrix expectedExcessReturns = new BigMatrix(a);
>
> double riskFreeReturn = 0;
> for(int i=1;i<=col;i++)
> riskFreeReturn+=returns[row-1][i-1];
> riskFreeReturn/=row;
>
> for(int i=1;i<=row;i++){
>
> double tmp = 0;
> for(int j=1;j<=col;j++)
> tmp+=returns[i-1][j-1];
> tmp/=col;
>
> expectedExcessReturns = expectedExcessReturns.set(i-1, 0, tmp-
> riskFreeReturn);
> }
>
> return expectedExcessReturns;
> }
>
> public static void main(String[] args)
> {
> int assetNum=7;
> double[][] assets_return={
> {-1.5905837442343828E-4, -0.03062360801781757,
> -0.029857534032853142, -0.011811692726036832,
> -0.017972310602803136, 0.017338003502626997, 0.0},
> {-0.02757158006362653, -0.02562704471101405,
> -0.011751538891997735, -0.024915062287655786, -0.01684088269454123,
> 0.013585351447135364, 0.0},
> {-0.00699300699300693, -0.033802816901408676,
> -0.04675196850393671, -0.021166752710376546, -0.007911392405063583,
> 0.03827751196172254, 0.0},
> {-0.007626310772164015, 0.0038424591738713027,
> 0.02488038277511978, 0.025210084033613675, -0.02003642987249557,
> -0.09758364312267642, 0.0},
> {-0.03965053763440893, 0.021693491952414375,
> 0.01643835616438392, -0.007412398921833087, 0.01765105227427014,
> -0.010006671114076025, 0.0},
> {-0.017821782178217872, 0.005040322580645311,
> 0.006018054162487363, 9.008107296569024E-4, 0.002999999999999824,
> -0.01196410767696908, 0.0},
> {2.630552127527583E-4, 2.5867028174649627E-4,
> 2.3866431891514327E-4, 1.9564035993080523E-4, 2.351016690966669E-4,
> 1.9070675120065465E-4, 0.0}
> };
>
> TestMarkowitz tm = new TestMarkowitz();
> BasicMatrix covariances = tm.getACovariances(assets_return);
> BasicMatrix expectedExcessReturns = tm.getAnExpectedExcessReturns
> (assets_return);
> BigDecimal riskAversion = new BigDecimal(1.0);
>
> MarketEquilibrium marketEquilibrium = new MarketEquilibrium
> (covariances,riskAversion);
> MarkowitzModel markowitzModel = new MarkowitzModel
> (marketEquilibrium,expectedExcessReturns);
>
> for(int i=0;i<assetNum;i++){
> markowitzModel.setLowerLimit(i, new BigDecimal(0.0));
> markowitzModel.setUpperLimit(i, new BigDecimal(1.0));
> }
> List<BigDecimal> re = markowitzModel.getWeights();
>
> System.out.println("=======result====================");
> for(int i=0;i<re.size();i++)
> System.out.println(re.get(i));
>
> return;
> }
>
> }
>
> /
> **********************************************************************
> **********/
>
> and the result turn out as:
>
> -0.25
> -0.25
> -0.25
> -0.25
> 0.75
> 0.75
> 0.75
>
> /
> **********************************************************************
> ******/
>
> If you find out any problem, just let me know , thanks a lot!
>
> On Thu, 15 Jan 2009 09:20:52 +0100, Anders Peterson wrote
> > You have to send code (snippets) that I can execute.
> >
> > Are you using v25 or the latest from cvs?
> >
> > How do you instantiate the ActiveSetSolver?
> >
> > What's the correct solution?
> >
> > You say that if you change the sign of "C" you get an ok result.
> > What does this mean? Did you make a mistake with the sign, or do
> > you suspect there is a bug in ojAlgo?
> >
> > I need better info regarding what you do, and what the problem is!
> >
> > /Anders
> >
> > On 15 jan 2009, at 06.14, Ruan Weichao wrote:
> >
> > >
> > > Thank you for your response.
> > >
> > > Here's my model's detail:
> > >
> > > [AE]={1.0, 1.0, 1.0, 1.0, 1.0, 1.0,
> 1.0}
> > >
> > >
> > > [be]={1.0}
> > >
> > >
> > > [Q]={3.048907897157133E-4, 1.6671472561019247E-4,
> > > 4.4500080981934345E-4, -5.389129745055723E-4,
> > > -2.6090705011393183E-4, -1.2633284900760366E-4,
> > > -6.485428846447651E-7},
> > > {1.6671472561019247E-4, 2.341985572849691E-4,
> > > 2.9113916450678265E-4, -4.5760873539850514E-4,
> > > 1.3078636134987255E-5, -2.354289901013046E-5,
> -7.578030042426654E-7},
> > > {4.4500080981934345E-4, 2.9113916450678265E-4,
> 7.46023915996829E-4,
> > > -0.0010247176498305568, -2.6745504327902895E-4,
> > > -1.6563544154823496E-4, -8.293698990696063E-7},
> > > {-5.389129745055723E-4, -4.5760873539850514E-4,
> > > -0.0010247176498305568, 0.001754169535149865,
> > > 2.0293065310212377E-4, 2.1401092557826588E-4,
> 1.0252846778608953E-7},
> > > {-2.6090705011393183E-4, 1.3078636134987255E-5,
> > > -2.6745504327902895E-4, 2.0293065310212377E-4,
> > > 4.632320892679136E-4, 1.7969731066037214E-4,
> 2.4953495129362833E-8},
> > > {-1.2633284900760366E-4, -2.354289901013046E-5,
> > > -1.6563544154823496E-4, 2.1401092557826588E-4,
> > > 1.7969731066037214E-4, 8..346410612364995E-5,
> -7.02099350897589E-8},
> > > {-6.485428846447651E-7, -7.578030042426654E-7,
> > > -8.293698990696063E-7, 1.0252846778608953E-7,
> > > 2.4953495129362833E-8, -7.02099350897589E-8,
> 8.367244992498656E-9}}
> > >
> > > [c]={{0.010638291263564232},
> > > {0.013500370827906071},
> > > {0.011390037735101773},
> > > {0.010385042339767682},
> > > {3.812208389845893E-4},
> > > {0.002315505853720011},
> > > {0.0}}
> > >
> > >
> > > [AI]={{1.0, 0.0, 0.0, 0.0, 0.0,
> 0.0, 0.0},
> > > {0.0, 1.0, 0.0, 0.0, 0.0, 0.0, 0.0},
> > > {0.0, 0.0, 1.0, 0.0, 0.0, 0.0, 0.0},
> > > {0.0, 0.0, 0.0, 1.0, 0.0, 0.0, 0.0},
> > > {0.0, 0.0, 0.0, 0.0, 1.0, 0.0, 0.0},
> > > {0.0, 0.0, 0.0, 0.0, 0.0, 1.0, 0.0},
> > > {0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 1.0},
> > > {-1.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0},
> > > {0.0, -1.0, 0.0, 0.0, 0.0, 0.0, 0.0},
> > > {0.0, 0.0, -1.0, 0.0, 0.0, 0.0, 0.0},
> > > {0.0, 0.0, 0.0, -1.0, 0.0, 0.0, 0.0},
> > > {0.0, 0.0, 0.0, 0.0, -1.0, 0.0, 0.0},
> > > {0.0, 0.0, 0.0, 0.0, 0.0, -1.0, 0.0},
> > > {0.0, 0.0, 0.0, 0.0, 0.0, 0.0, -1.0}}
> > >
> > > [bi]={{1.0},
> > > {1.0},
> > > {1.0},
> > > {1.0},
> > > {1.0},
> > > {1.0},
> > > {1.0},
> > > {0.0},
> > > {0.0},
> > > {0.0},
> > > {0.0},
> > > {0.0},
> > > {0.0},
> > > {0.0}}
> > >
> > > And here's the result:
> > >
> > > -0.25
> > > -0.25
> > > -0.25
> > > -0.25
> > > 0.75
> > > 0.75
> > > 0.75
> > >
> > > And i found that if i change the nagetive ExceptedReturn to
> > > positive, the result will be fine.
> > >
> > >
> > > On Wed, 14 Jan 2009 09:27:53 +0100, Anders Peterson wrote
> > > > How small are the negative values?
> > > >
> > > > Give me a unit test (something that I can easily integrate
> with the
> > > > existing junit tests) that shows what you do, and where/how you
> > > > think it goes wrong.
> > > >
> > > > /Anders
> > > >
> > > > On 14 jan 2009, at 03.58, Ruan Weichao wrote:
> > > >
> > > > >
> > > > > Hi,
> > > > >
> > > > > I use the Markowitz model to calculate the optimize some
> assets'
> > > > > weights,
> > > > >
> > > > > and set all the assets' lowerlimit with Zero and set all the
> > > > > upperlimits with
> > > > >
> > > > > One, but it turns out with some negative value which are
> little
> > > > > than Zero, it
> > > > >
> > > > > seems that the lowerlimit does not work.
> > > > >
> > > > > So I want to ask what should I do it to confirm the
> lowerlimit?
> > > > >
> > > > > --
> > > > > Best regards
> > > > >
> > > > > weichao
> > > > >
> > > > >
> > >
> ----------------------------------------------------------------------
> > > > > --------
> > > > > This SF.net email is sponsored by:
> > > > > SourcForge Community
> > > > > SourceForge wants to tell your story.
> > > > > http://p.sf.net/sfu/sf-spreadtheword
> > > > > _______________________________________________
> > > > > ojAlgo-user mailing list
> > > > > ojA...@li...
> > > > > https://lists.sourceforge.net/lists/listinfo/ojalgo-user
> > > > >
> > > > >
> > > >
> > > >
> > >
> ----------------------------------------------------------------------
> > > ------
> > > --
> > > > This SF.net email is sponsored by:
> > > > SourcForge Community
> > > > SourceForge wants to tell your story.
> > > > http://p.sf.net/sfu/sf-spreadtheword
> > > > _______________________________________________
> > > > ojAlgo-user mailing list
> > > > ojA...@li...
> > > > https://lists.sourceforge.net/lists/listinfo/ojalgo-user
> > >
> > >
> > > --
> > > Best regards
> > >
> > >
> ----------------------------------------------------------------------
> > > --------
> > > This SF.net email is sponsored by:
> > > SourcForge Community
> > > SourceForge wants to tell your story.
> > > http://p.sf.net/sfu/sf-
> > > spreadtheword_______________________________________________
> > > ojAlgo-user mailing list
> > > ojA...@li...
> > > https://lists.sourceforge.net/lists/listinfo/ojalgo-user
> >
> >
> ----------------------------------------------------------------------
> ------
> --
> > This SF.net email is sponsored by:
> > SourcForge Community
> > SourceForge wants to tell your story.
> > http://p.sf.net/sfu/sf-spreadtheword
> > _______________________________________________
> > ojAlgo-user mailing list
> > ojA...@li...
> > https://lists.sourceforge.net/lists/listinfo/ojalgo-user
>
>
> --
> Best regards
>
> ----------------------------------------------------------------------
> --------
> This SF.net email is sponsored by:
> SourcForge Community
> SourceForge wants to tell your story.
> http://p.sf.net/sfu/sf-
> spreadtheword_______________________________________________
> ojAlgo-user mailing list
> ojA...@li...
> https://lists.sourceforge.net/lists/listinfo/ojalgo-user
|
