Simulating ages

[duskstar]

This isnt so much a Stata issue as it is a statistical one. I am creating a simulation ahnd I want to simulate some plausible ages. What is the best distribution to sample from? I had assumed the normal distribution, but if so, is there anyway to restrict it in some way (like a half normal) to stop it from sampling negative values?

[Dason]

Well it depends on what you're simulating ages for. If you're simulating a college aged group that will be different than simulating ages for people in retirement homes.

[duskstar]

I might have solved this. I took a dataset similiar to what I wanted to create and took the mean and SD and used that to simulate from a normal distribution.

I didnt want any particular subset of ages, I wanted patients of all ages, though there were more elderly people than younger.

[Dason]

Well if you have a dataset similar to what you want why not just sample from those ages with replacement?

[duskstar]

Well if you have a dataset similar to what you want why not just sample from those ages with replacement?

I hadn't thought of that, I will try that. Thanks for the suggestion.