Regression Analysis II: Linear Models

Instructor(s):

• Timothy McDaniel, Statistics, Buena Vista University
• Brian Pollins, Political Science, Ohio State University

The prerequisites for the workshop are an introductory course in applied statistics at the level of Neter et al., Applied Statistics, and a background in elementary mathematics sufficient for the study of matrix algebra. The content of this course will include the nature of a linear model, least squares and maximum likelihood estimation, analysis of residuals, the general linear model, violation of assumptions (multicollinearity, heteroscedasticity, autocorrelation, measurement error, specification error), models with dummy variables, analysis of variance, and analysis of covariance. Although knowledge of matrix arithmetic is not a prerequisite for this course, some concepts in matrix algebra will be introduced as appropriate. Wonnacott and Wonnacott's Econometrics, Neter and Wasserman's Applied Linear Statistical Models, and Weisberg's Applied Regression Analysis are three of a large number of texts that could be used for this course.

Dates:  June 22-July 17 

Fee:  consult the fee structure

Dates:  July 20-August 14 

Fee:  consult the fee structure