Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: How to compute with Stata probabilities by conditioning ?


From   "Tiago V. Pereira" <tiago.pereira@mbe.bio.br>
To   statalist@hsphsun2.harvard.edu
Subject   st: How to compute with Stata probabilities by conditioning ?
Date   Mon, 20 May 2013 12:36:39 -0300 (BRT)

Many thanks, Roger! That makes sense.

Cheers!

Tiago


If you type, in Stata,

help density_functions##normal

then you will find that the function binormal(h,k,r) computes the joint
cumulative distribution function, not the conditional diatribution
function as you seem to believe. The .pdf documentatio gives the equation
explicitly.


I hope this helps.

Best wishes

Roger
--
Dear statalisters,

Imagine two standard normal variables, X and Y, with known correlation, rho.

P(X>t1|Y>t2) can be computed with

binormal(t1,t2,rho)

What if we were interest in computing

P(X>t1 and Y>t2) ?

Since X and Y are correlated, I am a bit confused.

I am using brute force (dumb way):

set obs 1000000000
matrix M = (1,rho \ rho, 1)
drawnorm x y, means(0 0) sds(1 1) corr(M)
count if x>t1&y>t2

P(X>t1 and Y>t2)  = r(N)/_N

Is there a smarter way of computing probabilities by conditioning (with
Stata)?

Thanks in advance.

Tiago





*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index