Linear Regression Analysis

A 3-Day Seminar Taught by Paul Allison, Ph.D. 


Linear regression is the most widely-used method for the statistical analysis of non-experimental (observational) data. It’s also the essential foundation for understanding more advanced methods like logistic regression, survival analysis, multilevel modeling, and structural equation modeling. Without a thorough mastery of linear regression, there’s little point in trying to learn more complex regression methods.

If you’ve never had a course on linear regression, or if you took one so long ago that you have forgotten most of it, this seminar will get you up to speed. In three days, we’ll cover almost a semester’s worth of material. When it’s over, you’ll be a knowledgeable and effective user of regression methods. And you will have the necessary preparation to take most of Statistical Horizons’ more advanced seminars.

Paul Allison has been teaching courses on linear regression for more than 30 years. He is the author of the popular text, Multiple Regression, which provides a very practical, intuitive, and non-mathematical introduction to the topic of linear regression.

The seminar will begin by focusing on the two major goals of linear regression: prediction and hypothesis testing. We’ll look at several examples from published articles to see how linear regression is used in practice and how to interpret regression tables.

Next we’ll consider all the things that can go wrong when using linear regression, and we’ll see how to critique the analyses done by others.

We’ll delve into the mathematical theory behind linear regression, focusing on the essential assumptions, and on the implied properties of the least squares method. We’ll also spend considerable time on techniques for building non-linearity into linear regression by way of transformations, interactions, and dummy (indicator) variables.

There will be lots of hands-on exercises using either SAS or Stata.


COMPUTING

This seminar will use both SAS and Stata for the many empirical examples and the exercises. At least one hour each day will be devoted to hands-on exercises. To optimally benefit, you should bring your own laptop with a recent version of SAS or Stata installed. Power outlets will be provided at each seat.

If you are unable to obtain access to the full versions of SAS or Stata, there is an option to obtain a trial version of Stata 13. Stata is licensed through StataCorp (www.stata.com) and is frequently offered at a significant discount through academic institutions to their employees and students. Seminar participants who are not yet ready to purchase Stata could take advantage of StataCorp’s 30-day software return policy and obtain Stata 13 on a trial basis in the weeks immediately preceding this course.  Stata also has a 30-day trial-license “share” policy permitting current license-holders to share a trial copy: http://www.stata.com/customer-service/share-stata/


Who should attend?

This seminar is designed for people who have a basic background in statistics, and who want to learn more about the theory and practice of linear regression. You’ll need to have taken an introductory course in statistics, and be comfortable with such concepts as random sampling, measures of center and variability, correlation, sampling distributions, standard errors, confidence intervals, and hypothesis testing. You should also have at least some experience using either SAS or Stata. Neither matrix algebra nor calculus will be used. Although the course is relatively non-mathematical, considerable emphasis will be placed on the underlying assumptions and their implications. Upon completion of this seminar, you should be able to run your own linear regressions, build and evaluate regression models, and interpret and critique regression results.


Materials

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


Registration and lodging

The fee of $1,295 includes all seminar 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 $142 per night. This location is about a 5-minute walk to the seminar location. To make reservations, call 203-905-2100 during business hours and identify yourself by using the code STA608. For guaranteed rate and availability, you must reserve your room no later than May 8, 2014.


SEMINAR OUTLINE 

  1. What is linear regression and what is it good for?
  2. Examples of published regression analyses and interpretation of results.
  3. The mechanics of regression in SAS and Stata.
  4. Bivariate and trivariate regression.
  5. Assumptions of linear regression and properties of least squares estimation.
  6. Evaluation of regression models.
  7. What can go wrong in linear regression.
  8. Regression, correlation, and standardized coefficients.
  9. Nonlinearity and interaction.
  10. Dummy (indicator) variables.
  11. Multicollinearity.
  12. Model building strategies.
  13. Missing data.
  14. Heteroscedastity.