Re: [ojAlgo-user] Problem with the constraints on the assets with Markowitz model

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

Your model is very artificial/symmetric…

The problem is that you've mixed up the lower and upper limits on the variables (asset weight constraints).

/Anders


On 27 okt 2011, at 15:10, Samuel Le wrote:

> Indeed you are right!
> My correlation matrix was wrong.
> Now I just slightly change the way I set the volatilities of the assets:
> 
> Double[] vols = new Double[n];
> for(int i = 0; i < n; i++) {
>        vols[i] = new Double(i+1)/ new Double(n);
> }
> 
> (the volatility of asset i is now (i+1)/n, so there are no zeros anymore).
> 
> Now the optimization stated is still FAILED. Also when I re-run the same code I get different weights. I will change my code to always check the optimisation state and ignore the weights provided when it's FAILED.
> Also this time I have no idea what causes the optimization to fail.
> 
> Samuel
> 
> -----Original Message-----
>> From: Anders Peterson [mailto:an...@op...]
> Sent: 27 October 2011 13:59
> To: oja...@li...
> Subject: Re: [ojAlgo-user] Problem with the constraints on the assets with Markowitz model
> 
> It seems I was lucky that one time. Running the code again I get solutions similar to yours.
> 
> The problem is your covariance matrix. It has columns and rows with all zeros. That's not allowed!
> 
> Also you can/should check the optimization state:
> 
> markowitzModel.getOptimisationState();
> 
> I get state "FAILED".
> 
> /Anders
> 
> 
> On 27 okt 2011, at 14:24, Anders Peterson wrote:
> 
>> Just ran your code. I got this solution:
>> 
>> {{0.05},
>> {0.05},
>> {0.05},
>> {0.05},
>> {0.05},
>> {0.05},
>> {0.05},
>> {0.05},
>> {0.3},
>> {0.3}}
>> 
>> I am of course using v31, but I don't know what would be the difference. Upgrade to v31 and see if the problem disappears for you as well.
>> 
>> /Anders
>> 
>> 
>> On 27 okt 2011, at 13:21, Samuel Le wrote:
>> 
>>> Dear all,
>>> 
>>> I am new to ojalgo and am trying to use it for portfolio optimisation.
>>> I think I found a bug: when trying to find some optimal weights with constraints, I can obtain inconsistent weights.
>>> I include below an example. I am using the ojalgo version 30.6.
>>> I would be grateful if someone could tell me what I am doing wrong.
>>> 
>>> Many thanks,
>>> Samuel
>>> 
>>> Here is my code:
>>> 
>>> import java.math.BigDecimal;
>>> 
>>> import org.ojalgo.finance.portfolio.MarketEquilibrium;
>>> import org.ojalgo.finance.portfolio.MarkowitzModel;
>>> import org.ojalgo.matrix.BasicMatrix;
>>> import org.ojalgo.matrix.PrimitiveMatrix;
>>> import org.ojalgo.matrix.store.PhysicalStore;
>>> import org.ojalgo.matrix.store.PrimitiveDenseStore;
>>> 
>>> public class OjalgoTest {
>>> 
>>>     public static void main(String[] args) {
>>>           int n = 10;
>>>           PhysicalStore<Double> resStore = PrimitiveDenseStore.FACTORY.makeZero(n, n);
>>>           Double[] vols = new Double[n];
>>>           for(int i = 0; i < n; i++) {
>>>                 vols[i] = new Double(i)/ new Double(n);
>>>           }
>>>           Double corr = 0.1;
>>>           for(int i = 0; i < n; i++) {
>>>                 for(int j = 0; j < n; j ++) {
>>>                       if(i == j) {
>>>                             resStore.set(i, j, vols[i]*vols[i]);
>>>                       }
>>>                       else {
>>>                             resStore.set(i,j,corr*vols[i]*vols[j]);
>>>                       }
>>>                 }
>>>           }
>>>           double[] expectedRets = new double[]{0.2,0.19,0.18,0.17,0.16,0.15,0.14,0.13,0.12,0.11};
>>>           PhysicalStore<Double> physicalStorDbl = PrimitiveDenseStore.FACTORY.makeZero(n, 1);
>>>           for(int i = 0; i < n ; i++) {
>>>                 physicalStorDbl.set(i, 0, expectedRets[i]);
>>>           }
>>>           PrimitiveMatrix expectedReturnsMatrix = new PrimitiveMatrix(physicalStorDbl);
>>>           PrimitiveMatrix covMat = new PrimitiveMatrix(resStore);
>>>           MarketEquilibrium me = new MarketEquilibrium(covMat);
>>>           MarkowitzModel markowitzModel = new MarkowitzModel(me, expectedReturnsMatrix);
>>>           for(int i = 0; i < n; i++) {
>>>                 markowitzModel.setUpperLimit(i, new BigDecimal("0.05"));
>>>                 markowitzModel.setLowerLimit(i, new BigDecimal("0.3"));
>>>           }
>>>           markowitzModel.setRiskAversion(new BigDecimal("0.0001"));
>>>           BasicMatrix tmpWgts = markowitzModel.getAssetWeights();
>>>           System.out.println(tmpWgts);
>>> 
>>>     }
>>> }
>>> 
>>> And this is what I am getting:
>>> 
>>> {{0.05},
>>> {0.05},
>>> {0.05},
>>> {0.05},
>>> {0.05},
>>> {0.05},
>>> {0.05},
>>> {0.05},
>>> {76.9098},
>>> {-76.3098}}
>>> 
>>> 
>>> __________ Information from ESET NOD32 Antivirus, version of virus signature database 6275 (20110707) __________
>>> 
>>> The message was checked by ESET NOD32 Antivirus.
>>> 
>>> http://www.eset.com
>>> ------------------------------------------------------------------------------
>>> The demand for IT networking professionals continues to grow, and the
>>> demand for specialized networking skills is growing even more rapidly.
>>> Take a complimentary Learning@Cisco Self-Assessment and learn
>>> about Cisco certifications, training, and career opportunities.
>>> http://p.sf.net/sfu/cisco-dev2dev_______________________________________________
>>> ojAlgo-user mailing list
>>> ojA...@li...
>>> https://lists.sourceforge.net/lists/listinfo/ojalgo-user
>> 
> 
> 
> ------------------------------------------------------------------------------
> The demand for IT networking professionals continues to grow, and the
> demand for specialized networking skills is growing even more rapidly.
> Take a complimentary Learning@Cisco Self-Assessment and learn
> about Cisco certifications, training, and career opportunities.
> http://p.sf.net/sfu/cisco-dev2dev
> _______________________________________________
> ojAlgo-user mailing list
> ojA...@li...
> https://lists.sourceforge.net/lists/listinfo/ojalgo-user
> 
> 
> __________ Information from ESET NOD32 Antivirus, version of virus signature database 6275 (20110707) __________
> 
> The message was checked by ESET NOD32 Antivirus.
> 
> http://www.eset.com
> 
> 
> 
> __________ Information from ESET NOD32 Antivirus, version of virus signature database 6275 (20110707) __________
> 
> The message was checked by ESET NOD32 Antivirus.
> 
> http://www.eset.com
> 
> 
> ------------------------------------------------------------------------------
> The demand for IT networking professionals continues to grow, and the
> demand for specialized networking skills is growing even more rapidly.
> Take a complimentary Learning@Cisco Self-Assessment and learn 
> about Cisco certifications, training, and career opportunities. 
> http://p.sf.net/sfu/cisco-dev2dev
> _______________________________________________
> ojAlgo-user mailing list
> ojA...@li...
> https://lists.sourceforge.net/lists/listinfo/ojalgo-user
> 
>