Differential Evolution

Hello guys : I am testing the differential evolution class: I have a very large basket (a qquadratic problem) that I want to optmize subject to contraints: equal, greater and less than.

When I run the optimization using other methods (no diffwerential evoution) I get convergence but it can take 30 minutes to run for 1400 tickers.

I want to give Genetic Algo a try to find optimal value opf weights. basically I want to know how to set up the problem correctly, please point me in the right direction.

what do I have to correct i the definition below? ——————————————————————————————– QPProblem problem = new QPProblem(F, equal, greater, less);

DEOptim solver = new DEOptim(new DEOptim.NewCellFactory() {
@Override
public DEOptimCellFactory newCellFactory() {
return new IntegralConstrainedCellFactory(new Rand1Bin(0.5, 0.5, uniform), new IntegralConstrainedCellFactory.AllIntegers());
}
}, uniform, 0.00001, 400000, 100);

IterativeSolution soln = solver.solve(problem);

————————————————————————–

I wonder if you could be so kind to point me out in the right direction.

How can I define the problem in suach a way that that the minumum is a vector of real numbers respecting the inequalitiers (equal, less and greater)

Do u have more examples illustrating the differential evolution class in algoquant? (besides the integer optimization of cross over strategy)

Thank you, ja