How can I compute the Multivariate normal cumulative distribution from vector a to vector b?

The function mvncdf(X) gives me the following probability: Pr{V(1) ≤ X(1),V(2) ≤ X(2),...,V(d) ≤ X(d)}

So it basically consider the the semi-infinite rectangle with upper limits defined by X.

However I want to consider a finite rectangle with upper limits defined by X and lower limits defined by Y.

I am looking for a function so that function(X,Y) gives the following probability:

Pr{ Y(1) ≤ V(1) ≤ X(1), Y(2) ≤ V(2) ≤ X(2),..., Y(d) ≤ V(d) ≤ X(d)}

Thank you. Marco

Answer 1

[Tom Lane]

Try "help mvncdf" again and read down to the syntax

Y = mvncdf(XL,XU,MU,SIGMA)

This sounds like what you want.

[Marco]

Thanks!