Statistics 850, Theory and Application of Regression and Analysis of Variance IIInstructor: Mary Lindstrom, Associate Professor, Department of Biostatistics andMedical InformaticsOffice Hours: To be announced and by appointmentOffice: K6/432 CSC (New hospital, North entrance, 4th floor)Phone: 262-4812Email: [email protected] or [email protected]: http://www.biostat.wisc.edu/faculty/lindstrom.htmlTeaching Assistant: Jialiang LiOffice Hours: To be announced and by appointmentOffice: 1205A MSCPhone: 263-7310Email: [email protected]: http://www.stat.wisc.edu/%7EjialiangLectures: T, Th 1:00 – 2:15, 101 PsycologyDiscussion Sections: #311: F 1:00 – 2:15; #312: F 2:25 - 3:40, 5295 MSCTextbook: Design a nd Analysis of Experiments, Dean and Voss, 2000ObjectivesThe primary objectives of this course are:• The practical and theoretical understanding of linear models at the graduate level, focusing on designedexperiments.• The use of the packages R and SAS to fit linear models and interpreting the output.• Handling “Messy data”. I.e., data that do not follow the usual assumptions or that do not have a standarddesign.• Communicating conclusions drawn from a statistical analysis of data.PrerequisitesStatistics 849 is the prerequisite for this class. If you have not completed 849 please see the instructor. A calculatorwill be required for exams. A scientific calculator is not necessary. Students may not share calculators duringexams. Some homework assignments will require the use of the statistical packages R and SAS. You also may beasked to read and interpret R or SAS output on exams. Rather than waiting until you need them, get started nowlearning to use them if necessary. The course web page (see below) will include introductory information.Discussion SectionsDiscussion sections will begin meeting the second week of class.Course Web PageThe course web page will be available later this week at http://www.stat.wisc.edu/∼st850-1It will include all homework assignments, exam dates, and other information useful for the course including aschedule that I will fill in as we go.Academic HonestyYou are permitted to talk to other students, your teaching assistant, and me about homework. However, you maynot present other people’s work as your own. If you received direct assistance with a problem please include areference or foot note to that effect. It is not acceptable for one student to write a solution for another student tocopy. You must work independently during exams. You may not share calculators during the exams.ExamsThere will be two in-class midterm exams and a final exam which will consist of an in-class component and atake-home component. Make-up exams will only be allowed under extenuating circumstancesHomeworkHomework ass ignments will be given out on Tuesdays and collected the following Tuesday at the beginning of class.There will be an assignment approximately every other week. Your homework solutions should be written up witha word processor, although you can handwrite sketches and mathematical notation. Graphs of data should becompleted in SAS or R and can be attached as figures.Each solution should include a brief description of the problem (that may be paraphrased from the actualproblem) as well as the solution. Take care to see that your written homework solutions are clear and easy to read.Put your solutions to problems in the order that they are assigned. Late homework will be penalized.GradingThe course grade will be determined from the midterms, (35%) homework (25%) and the final exam (40%).Suggested Reading (on reserve at Wendt Library)The course text does not include all of the information covered in this class. There is no one book that does. Thefollowing either are or will be on reserve at We ndt library. You may decide to buy one or more of these in additionto the course text, especially if you feel you need more help with SAS or R.Cochran and Cox, Exp erime ntal Designs (2nd edition), 1992Excellent reference for various designs including fractional factorialsDean and Voss, Design and Analysis of Experiments, 2000Class textLittell, Milliken, Stroup and Wolfinger, SAS System for Mixed Models, 1996SAS manual for mixed modelsLittell, Stroup and Freund, SAS for linear models (4th edition), 2002SAS manual for linear modelsMcCulloch and Searle, Generalized, Linear, and Mixed Models (2nd edition), 2001The standard reference on Generalized linear modelsMuller and Fetterman, Regression and ANOVA, An Integrated Approach Using SAS Software, 2002SAS manual for linear modelsPinheiro and Bates, Mixed Effects Models in S and S-PLUS, 2000We will be using this reference for working with Mixed models in R.Rencher, Linear Models in Statistics, 2000Technical reference which covers both regression and ANOVA and the connection betwe en the two.Snedecor and Cochran, Statistical Metho ds, (8th edition), 1989Old but good reference .Venables and Ripley, Modern Applied Statistics with S (4th Edition), 2002Suggested reference for