BlueElement
New Member
- Joined
- Aug 17, 2015
- Messages
- 5
Hi there, I am trying to use Solver for portfolio optimization. I'm using 9 stocks and Solver runs fine when I try to optimize for the lowest monthly draw-down using all 9 stocks (it gives a weight to each stock from -1 to 1. Where I run into an issue is if I add a constraint that requires that 6 of the 9 have weights of 0, aka I want to find the best 3 stocks to use, but no more than this. I'm trying to use Solver to come up with the best hedges but I don't want to use all 9 stocks. This isn't a constraint issue as I can manually enter some weights and run it tells me that:
I am using the GRG Nonlinear Method and receive the result "Solver converged in probability to a global solution". If I use the Evolutionary method I receive a similar result : "Solver cannot improve the current solution. All Constraints are satisfied."
Yet I can easily change the weights manually and find a better solution. So Solver isn't really helping in this instance. Has anyone had a similar problem? Any ideas on how to fix?
I am using the GRG Nonlinear Method and receive the result "Solver converged in probability to a global solution". If I use the Evolutionary method I receive a similar result : "Solver cannot improve the current solution. All Constraints are satisfied."
Yet I can easily change the weights manually and find a better solution. So Solver isn't really helping in this instance. Has anyone had a similar problem? Any ideas on how to fix?
