STAT 700 Fall 2014Data Analysis Methods Professor BaileySyllabusCourse Web Page: http://rohan.sdsu.edu/∼babailey/stat700and blackboard.sdsu.eduMeeting Time: Lectures: MW 2:00 - 3:15 p.m. SSW 2649Instructor: Professor Barbara BaileyGMCS 513email: [email protected] Hours: M 5:15 - 6:15 p.m. and T 4:30 - 5:30 p.m.; by appointmentReference: The textbook fo r the co urse isWest, B. T., Welch, K. B., and Galecki, A. T. (2007). Linear Mixed Models: A Practical Guide UsingStatistical Software, Chapman Hall/CRC Press.Objectives: This course will provide students with basic theory and tools for advanced, computationallyintensive data analy sis techniques and applica tions. The technique areas include random and mixedeffects models, repea ted measures and longitudinal data analysis, generalized linear models, multilevelmodels, and nonlinear models.With the explosion of computing power, data analysis methods for handling massive amounts of messydata with intricate correlation and signal patterns has become the norm. The course will equip studentswith the knowledge to apply a nd communicate statistica l inference drawn from modern computationalintensive data analysis techniques used in statistical practice.Homework: Homework assignments will be regularly available on the course web page as announced inclass. The homework will contain a series of prac tice problems of which selected problems will begraded. The homework serves as a tool to review and practice the material covered in class . Allmaterial covered on the assignments can be questioned on the exams. Some problems may requir ecomputing and must include concise computer output with a clearly presented version of your code.Late homework will not be acce pted. You may drop your lowest percentage score.Exams: There will be one in-class midterm Monday October 13, with a take-home portion due approxi-mately the next week. The in-class part of the exam will be closed book. A hand ca lc ulator is necessaryfor all in exams. No collaboration of any kind is allowed on the take-home part of the exam.No makeup exams are given - no exceptions.The final ex am will be g iven Monday, December 15 from 1:00 p.m. to 3 :00 p.m. in SSW-2649. Thefinal will be cumulative and comprehensive.Project: As part of the course you will be asked to do an individual data analysis project. The projectgrade will be based in part on a brief 5 minute presentation (depending on the size of the class) duringthe last full week of cla sses and a brief 3-5 page written report in journal style forma t (i.e., 12 pt font,one inch margins, single-spaced, figures and tables clearly pr e sented and lab e led at the end of theabstract, page limit does not include figures , tables, nor bibliography).You are required to attend all project prese ntations. Attendance at the presentations will be a part ofyour pr oject grade.The project will be done individually. You will illustrate and present data analysis concepts from theclass or literature. In c onsultation with me, you may may choose a project of inter e st to you. As part1of the project, expect to read the appropriate literature, write a report, and give an oral pre sentationto demonstrate a thorough understanding of and to illustrate the techniques/methods used in the classand article.Grading: The grade for the clas s is based on a score composed of the following.Homework 30 %Midterm Exam 30 %Project 10 %Final Exam 30 %Topics to be covered: basic outline; topics may be a dded and/o r dropped as the semester proceeds.1. Review Linear Models and ANOVAa. Matrix Theoryb. Estimation and Projectionsc. Quadratic Formsd. Inference about Normal Models and Bootstrapping2. Linear Mixed Modelsa. Fixed Effects vs. Rando m Effectsb. Estimation of Parametersc. Computational Issuesd. Hypotheses Tes tinge. Model Building Strategiesf. Model Diagnostics3. Multilevel Models4. Models for Repeated-Measures Da ta5. Models for Longitudinal Data6. Nonlinear Modelsa. Nonlinear L e ast Squaresb. Nonlinear Mixed ModelsPrerequisites: STAT 551B or 670B and STAT 510Tardiness and Early exits: The class time is from 2:00 - 3:15 p.m. As common courtesy to your fellowstudents, we would apprecia te if you show up to class on time and leave when dismissed at 3:15. Ifyou must leave early, please inform me and sit on the aisle near an exit so as not to disturb studentslistening to and trying to learn from the lectures.Code of Academic Conduct on Examinations and Assignments: “At San Diego State University,students are invited to be active members of the educational community. As with any community, itsmembers serve a vital r ole in determining acceptable standards of conduct, which includes academicconduct that reflects the highest level of honesty and integrity.” The “Statement of Student Rightsand Responsibilities clarifies for students their role as members of the campus community, setting forthwhat is ex pected of them in terms of behavior and contributions to the success of our university.” “In-appropriate conduct by Students . . . is subject to discipline on all San Diego State University Campuses.The Center for Student Rights and Responsibilities coordinates the discipline process and establishesstandards and procedures in accorda nce with regulations contained in Sections 41301-41304 of Title5 of The California Code of Regulations, and procedures contained in Exe c utive Order 628, StudentDisciplinary Procedures for The California State University.” See http://www.sa.sdsu.edu/srr/judicialfor more info rmation.2Students with Disabilities: If you are a student with a disability and believe you will need a c c ommoda-tions for this class, it is your responsibility to contact Student Disability Services at (619) 594-6473.To avoid any delay in the receipt of your accommodations, you should contact Student DisabilityServices as soon as possible. Please note that ac c ommodations are not retro active, and that accom-modations based upon disability cannot be provided until you have presented your instr uctor with anaccommodation letter from Student Disability Ser vices. Your cooperation is appreciated.Other information: See course web page: