Multivariate quadrature (approximation of joint distribution for portfolio choice)

2 views (last 30 days)
I would like to numerically compute an optimal portfolio, using multiple assets, which are correlated.
So my question is:
  1. Is there a standard approach for multi-dimensional quadrature? (standard deviation and covariance are sufficient statistics). I only saw this on the file exchange: https://nl.mathworks.com/matlabcentral/fileexchange/13508-multi-dimensional-gauss-points-and-weights
  2. Or is the standard approach to use Monte Carlo simulation, using random draws from a multi-variate distribution (random number generator)
I specifically do not want to use theoretical solutions, but numerical ones.
Many thanks in advance!
  2 Comments
Torsten
Torsten on 16 Nov 2022
Edited: Torsten on 16 Nov 2022
The standard approach is to use "int" for symbolic integration or "integral", "integral2", "integral3" for numerical integration.
Sargondjani
Sargondjani on 17 Nov 2022
@Torsten thank you!! That works very nice, at least upto 3 dimensions... I guess for higher dimensions I'll have to stick with Monte Carlo simulation.

Sign in to comment.

Answers (0)

Categories

Find more on Portfolio Optimization and Asset Allocation in Help Center and File Exchange

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!