# st: RE: simulate binary data based on probability

 From "Nick Cox" To Subject st: RE: simulate binary data based on probability Date Wed, 2 Jul 2003 14:26:10 +0100

```fyzzz

> A problem in simulation torment me. Any suggestion is appreciated!
>
> I got a survey data (outcome: disease, yes/no), which is supposed to
> be too large in sample size. This could result in Type I
> Error. I try
> to solve this problem through simulating data.
>
> Based on the original data, I built a logistic model to obtain a
> fitted probability to each subject. And then, I want to simulate
> binary data for one particular health outcome (1 for yes
> and 0 for no)
> based on the fitted probability (from above step) using binomial
> probability distribution. But I don't know how to do this kind
> simulation is stata?

If I understand you correctly, this is the heart of
the matter:

Suppose probability of yes is 0.12345. Then

gen simulated = uniform() <= 0.12345

yields a variable with values 1 or 0 according
to whether random uniforms fall below or above
the stated threshold.

Nick
n.j.cox@durham.ac.uk

*
*   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/
```