5166 STA Statistics in Applications I FALL SEMESTER 2004
Instructor: Oce: Class Hours: Oce Hours: Text 1: Text 2: Xu-Feng Niu 210B OSB 644-4008 9:30 am - 10:45 am MW 11:00 am - 12:00Noon MW Box, G. E., Hunter, W. G. and Hunter, J. S (1978), Statistics for Experimenters, John Wiley and Sons, Inc. Spector, P. (1994), An Introduction to S and S-Plus, Wadsworth Publishing Company, Belmont, California.
Course Objectives: Statistics is an applied science whose focus is upon the collection, analysis and interpretation of data. Statistics is also a mathematically-based theoretical science in which the theory necessary to carry out data analysis is developed. This course will focus on the design of experiments which is the most valuable aspect of statistical method. Frequently conclusions are easily drawn from a well-designed experiment, even when rather elementary methods of analysis are employed. Conversely, even with the most sophisticated statistical analysis can not salvage a badly designed experiment. The topics of this course will include comparison of two and more means, random sampling, randomization and blocking with paired comparisons, statistical inference for means, variance, proportions and frequencies, and two-way factorial designs and Latin square design. The course will use the Splus as the statistical computing package. Introductory documents for the Statistics Departments network of Sun computers will be made available. student Each should have a good calculator. A computer account will be provided on the statistical department Sun system.
Prerequisites:
MAC 3312 or consent of the instructor
Course Grade:
Weights Homework 30% Midterm 30% Final 40% Grading 90-100 80-90 70-80 60-70 Below 60 Scale A B C D F
Homework: Please do your homework on standard size paper (8.5 11in). If there is more than one page, staple the pages together. In data analysis problems, do not turn in reams of computer output. Include only those parts of the output which are most relevant for a nal report. Tell in clear English the purpose of each step of the analysis, and tell what the plot or statistic shows.
Tentative Syllabus
Topic Introduction to the S language Some basic statistical concepts Random sampling, the sampling distribution of the sample mean (CLT) Basic notions in experimental design: Randomization and Blocking. Signicance tests and condence intervals for means, variances and proportions. The completely randomized design (One way layout) Methods of simultaneous inference Randomized blocks design and two-way factorial designs. Latin Squares design Factorial designs at two levels Applications of Factorial designs Fractional factorial designs at two levels
Source Chs. 1-6, An Introduction to S and S-Plus. BHH BHH Ch. 2. Ch. 3
BHH
Ch. 4
BHH
Ch. 5
BHH
Ch. 6
BHH BHH
Appendix 6C Ch. 7
BHH BHH BHH BHH
Ch. 8.1 and Appendix 8B Ch. 10 Ch. 11 Ch. 12