Rachel - Given the entire 4 x 4 covariance matrix and 4 x 1 mean vector,
you can show that the joint distribution of X2, X3 and X4 given X1 is
joint normal with mean vector
mu2 + S21(S11^-1)(X1 - mu1)
and covariance matrix
S22 - S21(S11^-1)S12
where mu2 is the 3 x 1 mean vector for (X2,X3, X4)' and mu1 = E(X1)
(scalar) and
where the entire covariance matrix is (S11 S12 \ S21 S22)
and where S11 is 1 x 1, S12 is 1 x 3, S21 is 3 x 1, and S22 is 3 x 3.
[for example, see Ch 2 in T. W. Anderson - Introducton to Multivariate
Statistics (Wiley) ]
So given X1, you can then generate the other three using drawnorm and
the above conditional mean and covariance matrix.
Al Feiveson
-----Original Message-----
From: [email protected]
[mailto:[email protected]] On Behalf Of Rachel
Sent: Wednesday, September 26, 2007 10:36 AM
To: [email protected]
Subject: Re: st: Drawing 3 normally distributed variable with a given
correlation structure to an existing variable
Hi Jeph,
-drawnorm- will draw 4 new variables. In my case, X1~N(0,1) is already
given, and I'm trying to draw X2,X3,X4~N(0,1) such that the four
variables have a given correlation structure.
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
