Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at

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

Re: st: drawnorm for discrete variables

From   John Antonakis <>
Subject   Re: st: drawnorm for discrete variables
Date   Fri, 30 Apr 2010 16:08:37 +0200


If you can't get Stata to do this, you might want to check out Mplus (, which can produce interesting combinations of variables in its Monte Carlo module. From the manual:

"Mplus has extensive Monte Carlo simulation facilities for both data
generation and data analysis. Several types of data can be generated:
simple random samples, clustered (multilevel) data, missing data, and
data from populations that are observed (multiple groups) or unobserved
(latent classes). Data generation models can include random effects,
interactions between continuous latent variables, interactions between
continuous latent variables and observed variables, and between
categorical latent variables. Dependent variables can be continuous,
censored, binary, ordered categorical (ordinal), unordered categorical
(nominal), counts, or combinations of these variable types. In addition,
two-part (semicontinuous) variables and time-to-event variables can be
generated. Independent variables can be binary or continuous. All or
some of the Monte Carlo generated data sets can be saved."



Prof. John Antonakis, Associate Dean Faculty of Business and Economics
Department of Organizational Behavior
University of Lausanne
Internef #618
CH-1015 Lausanne-Dorigny

Tel ++41 (0)21 692-3438
Fax ++41 (0)21 692-3305

Faculty page:

Personal page:

On 30.04.2010 15:57, Tyler Frazier wrote:
Hi, thank you for your response and observations.

I am looking for a relatively simple answer, which I plan to improve
upon in time.

The data will be used in an urban simulation which incorporates a
number of location choice models, regression, and other simple models.
 For now I would like to establish a method for synthetically
generating the data, for use in the larger spatial, framework.  At
this point, I am seeking a plausible solution for running the urban
simulation in order to observe its behavior.

I want to generate a population of approximately 200,000
persons/households. Ideally, the persons/households datasets for this
african city would have the following variables (see below).

As indicated, some of these variable are continuous (income), while
others are categorical (tribal affiliation) or binary (sex)

the survey itself is a fairly large sample, with about 3500 persons in
approximately 1100 households, for the applicable metropolitan area.
there is also a question of how to iteratively sample household and
person data

Any suggestions on the direction I should take are greatly appreciated.

Best regards,

. summarize

    Variable |       Obs        Mean    Std. Dev.       Min        Max
    hhrelate |      3447    2.714244    1.887394          1         10
         sex |      3447    1.506818    .5000261          1          2
         age |      3447    27.49956    17.14807          3         90
    ethnicit |      3291    22.25433    16.06378          1         90
    religion |      3445    5.190131    4.651516          1         96
    marstatu |      2808    3.746439    2.297089          1          6
    educatio |      3181    3.600126    3.494621          1         16
      income |      3444     6046706    2.21e+07          0   5.76e+08
    occupati |      1382    561.4363    245.7125         11        933
    industry |      1376    5334.578    2261.297        111       9900
      sector |       757    6.015852    2.695727          1         10

On Fri, Apr 30, 2010 at 3:39 PM, Maarten buis <> wrote:
--- On Fri, 30/4/10, Tyler Frazier wrote:
How to synthetically generate a population from a
sample where the variables are continuous, discrete
and binary?
The answer can range from very simple to very hard
(impossible), and it depends on what you want to
use the sample for and on what kind of information
you have to base your sampling on.

-- Maarten

Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen

*   For searches and help try:

*   For searches and help try:
*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index