STATISTICS 211 -- Statistical Methods II: Categorical Data Analysis

   instructor: Dan Gillen
       office: 346C Computer Science
        phone: 949-824-9862
        email: dgillen@uci.edu

Objectives: Introduction to statistical methods for analyzing 
            categorical data. Methods covered include ....

Prerequisites: Statistics 210 or equivalent.
Tentative Course Outline:

Weeks 1-2 - Modes of inference in the two sample case:  Randomization-based
	and model-based inference in the context of two-sample inference.  The
	planning of studies and the use of diagnostics to verify modeling 
	assumptions are also discussed.  (Reading ... Chapter 15, Appendix A.6-7)

Week 3 - Analysis of variance (ANOVA): One-way analysis of variance for 
	comparing more than two populations.  Issues discussed include: inference,
	power/sample size, multiple comparisons, contrasts, diagnostics.
      (Reading ... Chapters 16-18)

Week 4 - Pairing/blocking to reduce variance: Strategies for improving the
	efficiency of the analysis of variance.  (Reading ... Chapter 21)

Week 5 - Factorial experiments: Extension of ANOVA to handle experiments with
	more than one factor.   (Reading ... Chapters 19, 20, 23, 24)

Week 6 - Correlation and simple linear regression: Measures of association for
	two continuous variables and the use of linear regression to describe
	the relationship of two continuous variables.   (Reading ... Chapters 1-5)

Weeks 7-9 - Multiple linear regression: Statistical inference for the multiple
	linear regression model.  Emphasis on interpretation of model parameters,
	diagnosis of failures of model assumptions, and corrective actions.
      (Reading ... Chapters 6-11) 

Week 10 - Model building and model selection: Techniques for constructing linear
	regression models including transformations and model selection.  The 
	relationship of regression and ANOVA.   (Reading ... Chapters 6-11 (cont'd) 
      and Chapter 22)


Course Requirements: Students will be assigned problem sets, approximately 
	once in each week.   The problem sets will make use of the SAS statistical
	software package.  There will be midterm and final exams.  The grade 
	is determined by performance on this work as follows: homework
	(approx 30%), midterm exam (approx 30%), final exam (approx 40%).

Texts/Reading:  
Primary text: 
Applied Linear Statistical Models
M. H. Kutner, C. J. Nachtsheim, J. Neter, W. Li (5th edition).  McGraw Hill Irwin, 2005. 

Material will also be drawn from: 
The Statistical Sleuth
F. L. Ramsey and D. W. Schafer (2nd  edition). Duxbury, 2002.

Prerequisite is a course like Statistics 7 or material in:
The Basic Practice of Statistics
D. S. Moore (2nd edition). WH Freeman, 2000.

Introduction to the Practice of Statistics
D. S. Moore and G. P. McCabe (4th edition). WH Freeman, 2003.

Mind on Statistics 
J. M. Utts and R. F. Heckard (2nd edition). Thomson/Brooks Cole, 2004.

Return to Course homepage
Return to UCI Statistics page