Mixed logistic regression; A comparison with WinBUGS

ADMB Files
Code: logistic.tpl
Data: logistic.dat
Initial values: logistic.pin
All required files (DOS): logistic.zip
All required files (linux): logistic.tar.gz
Results: logistic.par
R (S-Plus) scripts: logistic.s

Running ADMB-executables
In a DOS window
Under linux

Command line options:
-l1 10000000 -l2 100000000
-l3 10000000 -nl1 10000000

WinBUGS files
Code (ascii text): bugs.txt
Data (ascii text): data.txt
Initial values (ascii text): inits.txt

Results: Computation times
ADMB-RE: 27 seconds.
WinBUGS: 700 seconds.

ADMB-RE home
Otter Research

Model description

Logistic regression is commonly used to relate a dichotomous response y to explanatory variables. This example use simulated data to examplify logistic regression with random effects. A full description of the model can be found here: logistic.pdf

Comparison with WinBUGS

WinBUGS is a software package for formulating and fitting Bayesian hierarchical models. From a user perspective WinBUGS and ADMB-RE appear similar because both software packages:
  • Allows models with random effects.
  • Contain a full programming language, and hence give the user great flexibility in formulating statistical models.
  • Are automatic, in the sense of hiding the technical details of computation from the user.
Despite these similarities, the underlying computational technology is very different. In this example we compare WinBUGS and ADMB-RE with respect to computation times on simulated data. The R (S-Plus) script used to generate data in the input format for both WinBUGS and ADMB-RE can be found here: logistic.s. You can modify this script to generate new datasets.