Logistic Regression Using Stata

A 2-Day Seminar on Regression Methods for Categorical Dependent Variables


Taught by Paul D. Allison, Ph.D.

Logistic regression is one of the most widely used methods in statistical analysis. In this seminar, you’ll learn virtually everything you need to know to become a skilled user of logistic regression. We’ll cover the theory and practice of binary logistic regression in great detail including topics such as

  • odds and odds ratios
  • maximum likelihood estimation
  • interpretation of coefficients
  • convergence failures
  • goodness of fit
  • contingency table analysis
  • response-based sampling

We’ll also cover more advanced topics including ordered logistic regression, multinomial logistic regression, discrete-choice analysis, and methods for analyzing longitudinal data (robust standard errors, GEE, fixed and random effects).

You’ll learn the basic syntax and many options for the following Stata commands: logit, logistic, probit, cloglog, ologit, mlogit, clogit, xtlogit, xtmelogit, xtgee, glogit, exlogistic, firthlogit.


Who should attend?

If you need to analyze categorical outcomes and have a basic statistical background, this course is for you. You should have a good working knowledge of the principles and practice of multiple regression, as well as elementary statistical inference. But you do not need to know matrix algebra, calculus, or likelihood theory.

All examples and exercises will use Stata. No previous knowledge of Stata is assumed, however. Furthermore, nearly all the techniques taught in the course can be translated fairly easily to other packages. Lecture notes using SAS are available to registrants upon request.

This is a hands-on course, so you should bring your own laptop loaded with a recent version of Stata. If you don’t currently have Stata but know a colleague who has perpetual license for Stata 12, you can arrange for a 30-day free trial by having your colleague fill out a form at http://www.stata.com/customer-service/share-stata/


Location, format, materials

The course meets on Monday, June 4, and Tuesday, June 5, at Temple University Center City campus, 1515 Market Street, Philadelphia, PA. It runs from 9 a.m. to 5 p.m. each day, with a one-hour break for lunch.

Participants receive a bound manual containing detailed lecture notes (with equations and graphics), examples of computer printout, and many other useful features. This book frees participants from the distracting task of note taking.


Registration and Lodging

The fee of $795 includes all course materials. 

Lodging Reservation Instructions
A block of guest rooms has been reserved at the Club Quarters Hotel, 1628 Chestnut Street, Philadelphia, PA at a special rate of $116 per night. This location is a short walk to the seminar location. In order to make reservations, call 203-905-2100 during business hours and identify yourself by using group code STA632. For guaranteed rate and availability, you must reserve your room no later than April 20, 2012.


Course outline

  1. Review of linear model
  2. Dichotomous dependent variables in linear regression
  3. Odds and odds ratios
  4. The logistic (logit) regression model
  5. Estimating the logit model with Stata.
  6. Details of maximum likelihood estimation
  7. Interpreting logit coefficients
  8. Generalized R-square and other measures of fit
  9. Factor variables
  10. Hypothesis tests
  11. Probit model and other link functions
  12. Nonconvergence of ML estimates
  13. Logit analysis for contingency tables
  14. Multinomial response models : unordered case
  15. Logistic models for ordered polytomies
  16. Latent variable interpreation
  17. Response-based sampling
  18. Longitudinal data
  19. Random effects models
  20. GEE estimation
  21. Fixed effects logit model
  22. Discrete choice models

Comments by recent participants

“The course presented what was described, only better than expected. I not only got a great review but also learned a lot and developed a better understanding of concepts and techniques I’ve been using. Dr. Allison’s organization of the material and the aspects he emphasized enhanced this experience. What I’ve learned will be immediately applicable to my job.” 
Kristina Jones, Pacific Institute for Research and Evaluation 

“Excellent lectures featuring interesting research examples! This course was a pleasure to take. Thank you.”
Diana Surdulescu, Charles River Associates 

“The course was excellent, very informative.”
Mukul Sonwalkar 

“Excellent overview for those that are using logistic regression on a regular basis.”
N. R. Payne, Children’s Hospitals and Clinics of Minnesota 

“Professor Allison is an awesome instructor. This course is one of my best learning experiences.”
Mansoo Yu, University of Missouri

“A very practical and well-paced course, even for persons for whom statistics is a new field.”
Adrian Vancea, Georgetown University 

“For someone who has basic theoretical understanding of logistic regression, this course fills in all the holes of applying theory to real-life applications. The insight of Dr. Allison is rare and would be a benefit for users of logistic regression at all levels of expertise.”
Charles Goodyear, Infoscitex 

“I have a longitudinal data set that is perfectly suited to logistic regression analysis, and after this class I realize that I haven’t even begun to unlock its potential. I can’t wait to get back to work and apply what I’ve learned.”
J.R. Keller, University of Pennsylvania