Description Usage Arguments Value Note Author(s) References
Estimation of multivariate GARCH models
1 2 | FitMultivariateGarch(returns, demean = 1, eps = 0,
df = 500)
|
returns |
: [matrix] (T x N) returns so rows must correspond to time and columns to assets |
demean |
: [scalar] specifies whether returns should be demeaned (if demean = 1) or not to estimate the model; default value is 1. |
eps |
: [scalar] used in enforcing a_ii + b_ii <= 1 - eps; the default value is zero |
df |
: [scalar] degree of freedom for the t-distribution; the default value is 500 to make it, basically, normal |
mu : [vector]
ATMF : [matrix] coefficient matrix A-tilde (in the notation of the paper)
BTMF : [matrix] coefficient matrix B-tilde (in the notation of the paper)
CTMF : [matrix] coefficient matrix C-tilde (in the notation of the paper)
Hhat : [matrix] forecasted conditional covariance matrix
Code for MATLAB initially written by Olivier Ledoit and Michael Wolf
Xavier Valls flamejat@gmail.com
A. Meucci - "Exercises in Advanced Risk and Portfolio Management" http://symmys.com/node/170, "E 136 - Equity market: multivariate GARCH process".
See Meucci's script for "FitMultivariateGarch.m"
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.