201A Syllabus

Stats 201A: Research Design, Sampling and Analysis

Syllabus (Fall 2008)

Homepage: http://www.stat.ucla.edu/~hqxu/stat201A/
Instructor: Hongquan Xu <hqxu AT stat.ucla.edu> Phone: (310) 206-0035
Lecture Hours: MWF 1-1:50pm BH 5273
Office Hours: MWF 2-2:50pm at MS 8955
TA: Gong Chen (Office Hours: Tuesday 12-1 and 2-3 at MS 8359)

Textbooks (1-2)/References (3-5):

D. C. Montgomery (2005). "Design and Analysis of Experiments," 6th ed., John Wiley. (5th ed. is okay)
S. K. Thompson (2002). “Sampling, Second Edition,” John Wiley.
C. F. J. Wu and M. Hamada (2000). “Experiments: Planning, Analysis and Parameter Design Optimization,” John Wiley.
S. L. Lohr (1999). "Sampling: Design and Analysis," Duxbury Press.
J. J. Faraway (2005). "Linear Models with R," Chapman & Hall.

Overview: We will spend six weeks on research design and four weeks on sampling. Topics include

Research Design: Basic principles, ANOVA, randomized block designs, factorial designs, blocking and confounding in factorial designs, fractional factorial designs, response surface methods, robust parameter design, experiments with random factors, and nested designs. [Chapters 1-8, 11-14 of Montgomery (2005)]
Sampling: Simple random sample, Confidence Intervals, Sample Size Estimation, Estimation of Various Other Parameters (Proportions, Ratios and Subpopulation Means), Unequal Probability Sampling (Hansen-Hurwitz and Hovitz-Thompson Estimators), Regression Estimation, Stratified Sampling, Cluster and Systematic Sampling, Multistage designs. [Chapter 1-6, 8, 11-13 of Thompson (2002)]

Grading

Homework 30%
Group Project 20%
Final Exam 50%

Group Project: due Thursday, December 11, 2008, 8:00am
You are required to design an expeirment, collect data and perform analysis. You are encouraged to work with other 1-2 students. The whole group hand in one written report.

Final Exam: Thursday, December 11, 2008, 8:00am-11:00am
The final exam is comprehensive and includes topics on research design, sampling and analysis.

Statistical Software:
You are expected to be familar with one statistical software (e.g., R, S+, SAS, Stata, etc.). You are expected to know how to perform linear regression and make various plots. I will use R commands and outputs in the lecture, but I won't teach how to use R. I will provide some sample R commands and outputs on the web.

R is a language and environment for statistical computing and graphics. You can download R and get information from the R Home Page (http://www.r-project.org/).