Tiago Pereira wrote:
I have to compute the statistical power of a test. To do that, I need to
generate several independent standard normal variables.
Let Zi; 1 =< i =< k denote independent standard normal variables, and k a
value ranging from 1 to 200. How do I generate 1000 independent standard
normal variables using Stata supposing that k=18?
--------------------------------------------------------------------------------
Consider using -generate-. The expectation for rho is zero, which might be
better for your purpose, if it's to do power analysis by simulation. If you
need to specify orthogonal correlation structure for a given set of
variables, then you would use -drawnorm-.
Joseph Coveney
clear
set more off
set seed 12345
set obs 1000
local k 18 // You can change this from 1 to 200.
forvalues i = 1/`k' {
generate double Z`i' = invnormal(uniform())
}
exit
*
* 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/
