Message-ID: <E9955391-CA6C-4655-850F-6B6364413DE3@u.nus.edu>
Date: 2016-01-21T09:49:15Z
From: Le Hoang Van
Subject: Multivariate student t distribution in rmgarch
Hi Alexios,
Is there a multivariate version of pdist (rugarch) that allows us to compute the joint probability of two returns, say, P(Rs < 0.01, Rt < 0.02) at a certain point of time? So I was trying to fit a DCC-GARCH model with multivariate student t innovations:
xspec = ugarchspec(mean.model = list(armaOrder = c(1,1)), variance.model = list(garchOrder = c(1,1), model = 'sGARCH'), distribution.model = 'norm')
uspec = multispec(replicate(2, xspec))
spec = dccspec(uspec = uspec, dccOrder = c(1, 1), distribution = 'mvt?)
fit = dccfit(spec, data = ret, fit.control = list(eval.se<http://eval.se> = TRUE))
I was able to retrieve the mean vector (fitted(fit)), the covariance matrix (rcov(fit)), and the shape parameter (rshape(fit)) but have no idea how to put them together to compute said probability. Thank you and look forward to your response.
Regards,
Van
[[alternative HTML version deleted]]