INSTRUCTOR. Prof. Irene Hueter, Department of Mathematics, the University of Massachusetts Boston, Office at Science Center 3-91, Phone (617) 287-6463, Fax (617) 287-6433, Email
irene_hueter@baruch.cuny.edu. For more information, visit my website at http://faculty.baruch.cuny.edu/ihueter/.OFFICE HOURS. Tuesdays 1:00-2:30PM, Thursdays 11:15AM-12:45PM, and by appointment.
AIMS OF THIS COURSE. The course aims to cover the basic statistical inference topics and some applied statistics topics. It is meant as a follow-up to a one-semester probability course and intended to convey the mathematical ideas along with the most recent computational techniques valuable to conduct data analyses. For the topics to be covered and an approximate schedule, see below.
AUDIENCE.
Statistics is all pervading in science and everyday life and determines
decisions but is also misunderstood. The enormous ongoing advances in computer technology
impact many fields and demand for new statistical methologies and innovation
in turning information into knowledge. Statistics also teaches us how much
credence should be given to whom to what and why and about when to take
chances. As Stephen Senn* writes
INSTRUCTIONAL FORMAT. Mostly lectures taught by the instructor. Everyone is strongly encouraged to participate in the class activities and ask questions.
TEXT. Probability and Statistics. The Science of Uncertainty, by Michael Evans and Jeffrey Rosenthal (Second Printing), W.H.Freeman and Co, New York, 2004, ISBN 0-7167-4742-1. The book can be ordered at www.amazon.com.
PREREQUISITES. MA 345, a one-semester probability course, or the permission of the instructor.
MATERIAL TO BE COVERED. We will cover Chapters 4, 5, 6, 7, 9, and 10 of the book, and if time permits, parts of Chapters 11 and 8. See below.
STATISTICAL DATA ANALYSIS. Data sets along with references and background can be found at http://lib.stat.cmu.edu/DASL/, Department of Statistics, Carnegie Mellon University. There are many statistical software packages and programming languages. I plan to use the language R. R is a free software package and available at CRAN - Network (http://cran.r-project.org/). For some help with R, click here. Its commercial version is S-Plus. At the command level, there are no essential differences between the two and the documentations are interchangeable.
HOMEWORK. Weekly homework will be assigned and collected. The homework will count towards your final course grade (40%). The lowest 30% of your homework sets will be dropped. Working the homework problems is essential to your success in this course and is one of the best ways to prepare for the exams. The students are encouraged to form groups outside class and work together. However, each student is asked to write up her/his homework solutions individually in her/his own words and turn them in separately. Be aware that we may not have much or any time to discuss homework problems in class. Instead, please use a few minutes before or after class and the office hours for that purpose. You will be able to download the homework assignments (click here).
EXAMS. There will be two exams (each worth 20% of your grade) and one comprehensive final exam (worth 20% of your grade). Important: For makeups, you must have proper written documentation, provided close in time to the exam to be missed, preferably before the exam. If you miss one exam, then your final exam will count double.
EXAM SCHEDULE.
SPRING VACATION. March 12 - 20, 2005.
OTHER IMPORTANT DAYS.
January 31: Add/drop ends.
February 21: Presidents' Day (Holiday).
March 21: Mid-semester.
April 7: Course withdraw deadline. Pass/fail deadline.
April 18: Patriots Day (Holiday).
May 11: Last day of classes.
May 16-20: Final examinations' week.
May 30: Memorial Day (Holiday).
ATTENDANCE. Excessive unexcused absences may lead to a failing grade.
GRADING. Your grade will be based on your performance in the homework (40%), the two semester exams (40%), and the final exam (20%). I also give 5-10% bonus credit for enthusiasm, continued active class participation, and extra efforts.
CELL PHONES. All cell phones should be turned OFF. Be aware that your cell phone at work is very disruptive to the class, your fellow students, and the instructor. If you expect an emergency call, please clear it with me before class.
ACCOMMODATIONS. Section 504 of the Americans with Disabilities Act of 1990 offers guidelines for curriculum modifications and adaptations for students with documented disabilities. If appicable, students may obain adaptation recommendations from the Ross Center for Disability Services at http://www.rosscenter.umb.edu/text/, M-1-401, (617-287-7430). The student must present recommendations and discuss them with each instructor within a reasonable period, preferably by the end of the Drop/Add period.
ACADEMIC HONESTY. This class falls under the University Policy on Academic Standards and Cheating, the University Statement on Plagiarism, the Documentation of Written Work, and the Code of Student Conduct as delineated in the catalog of Undergraduate Programs, pp. 44-45, and 48-52. Academic dishonesty is unacceptable, will not be tolerated, and entails sanctions since it undermines the University's educational mission and the students' personal and intellectual growth. Please inform yourself of the principles and rules at http://www.umb.edu/student_services/student_rights/code_conduct.html. Be aware that ignorance of the Code is not an excuse and that academic dishonesty may have severe consequences in your student career.
GETTING HELP. The following are crucial factors to your success in this class: attend class regularly, read upcoming sections ahead of time so that you are prepared to ask questions in class, start homework early, and get help when there are signs of difficulties, for instance, low homework and exam scores. You are strongly encouraged and very welcome to make frequent use of my office hours, to ask about solutions to homework problems in particular. While the office hours probably are the most reliable times to get a hold of me, you may also reach me at other times in my office (Tu, Wedn, Th), over the phone, or by sending me email.
APPROXIMATE SCHEDULE Week Topics Book Sections 4 - Sampling Distributions and Limits 1 (1/25) Introductory Lecture Modes of convergence, Central Limit theorem (CLT) 4.2, 4.3, 4.4 Monte Carlo approximations, Buffon needle problem 4.5 4 (cont'd) and 5 - Statistical Inference 2 (2/1) Chisquare-, t-, F-distributions 4.6 Statistical inference (5.1 for reading) 5.1, 5.2 Statistical models, data collection 5.3, 5.4 Descriptive statistics, plotting data, types of inferences 5.5 6 - Likelihood Inference 3 (2/8) Likelihood function, sufficient statistics 6.1 Maximum likelihood estimation (MLE) 6.2 Inferences based on MLE, confidence intervals 6.3 4 (2/15) Hypothesis testing, confidence intervals, power 6.3 Distribution-free methods, methods of moments, 6.4 Bootstrapping 5 (2/22) Large sample behavior of the MLE 6.5 Review (for exam 1) 7 - Bayesian Inference 6 (3/1) Prior and posterior distributions 7.1 EXAM 1 7 (3/8) Estimation 7.2 Credible intervals, hypothesis testing, Bayes factor 7.2 SPRING VACATION (March 12 - 20, 2005) 8 (3/22) Prediction 7.2 Bayesian computation, Gibbs sampling 7.3 7 (cont'd) and 9 - Model Checking 9 (3/29) Choosing priors 7.4 Review (for exam 2) Checking the sampling model 9.1 10 (4/5) EXAM 2 Bayesian model checking 9.2 Multiple checks 9.3 10 - Relationships Among Variables 11 (4/12) Related variables, cause-effect, design of experiments 10.1 Categorical response and predictors 10.2 Quantitative response and predictors 10.3 Method of least squares, simple linear regression Bayesian formulation 12 (4/19) Multiple linear regression 10.3 Quantitative response and categorical predictors 10.4 ANOVA (one-way and two-way), randomized blocks 11 - Markov Chains 13 (4/26) Random walks, Markov chains 11.1, 11.2 Markov chain Monte Carlo 11.3 11 (cont'd) and 8 - Optimal Inferences 14 (5/3) Metropolis-Hastings algorithm, Gibbs sampler 11.3 Optimal unbiased estimation, Cramer-Rao inequality 8.1 Optimal hypothesis testing, likelihood ratio tests 8.2 15 (5/10) Optimal Bayesian inferences 8.3 Decision theory 8.4