STATISTICS 225 (#60170) APPLIED REGRESSION ANALYSIS Summer 2013 ********************************************************** INSTRUCTOR: Mun Son, mson@uvm.edu Rm. 309, Math/Stat Dep't., 16 Colchester Ave. (The Lord House) 656-4329(O), 879-0721(R), 734-2521(CP) Office Hours: Monday ? Thursday: 4:00 ? 4:50 pm. GTF: Basement of Mansfield House, Votey parking lot, 656-3349(O) Office Hours: CLASS ROOM & CLASS HOURS: 5:00 pm - 8:45 pm TR, LAFAYE HALL L411 May 20, 2013 - June 28, 2013 TEXTBOOK: Applied Linear Regression Models by Kutner, Nachtsheim and Neter, 4th Ed., Irwin. ISBN: 0-07-238691-6 JMP, SAS/ETS --- recommended. COURSE OUTLINE: Regression analysis is a powerful, basic statistical technique which is being used in many fields. All statistical computing systems provide linear regression programs. The insight necessary to answer important questions such as - How can we find out if the regression assumptions are wrong? If they are, what do we do? Why is the order of entry of terms into the model important? Should we transform the response variables? If so, how? Should terms be dropped from a polynomial model? How can different batches of raw material be introduced in the regression equation? Are individual confidence intervals for regression coefficients good or bad? Why are coefficients tested with both t and F statistics? What is a general linear models? When should you use stepwise regression to build a model? What is nonlinear estimation? To obtain these goals, we will cover Chaps. 1 - 3, 5, Chaps. 6 - 11, and Chap. 14. Some sections will be skipped. Chap. 1 -- Linear Regression with One Predictor Variable Chap. 2 -- Inferences in Regression and Correlation Analysis Chap. 3 -- Diagnostics and Remedial Measures Chap. 5 -- Matrix approach to Simple Linear Regression Analysis Chap. 6 -- Multiple Regression I Chap. 7 -- Multiple Regression II Chap. 8 -- Regression Models for Quantitative and Qualitative Predictors Chap. 9 -- Building the Regression Model I: Model Selection and Validation Chap. 10 -- Building the Regression Model II: Diagnostics Chap. 11 -- Building the Regression Model III: Remedial Measures Chap. 14 -- Logistic Regression, Poisson Regression, and Generalized Linear Models Some sections will be skipped. GRADING: HW/Project - 10% Exam I - 40% (Tue., 6/11/13) Final - 50% (Thu., 6/27/13, Comprehensive) All students are expected to regulate their class attendance. STATISTICS 225 (#60170) APPLIED REGRESSION ANALYSIS Summer 2013 ********************************************************** INSTRUCTOR: Mun Son, mson@uvm.edu Rm. 309, Math/Stat Dep't., 16 Colchester Ave. (The Lord House) 656-4329(O), 879-0721(R), 734-2521(CP) Office Hours: Monday ? Thursday: 4:00 ? 4:50 pm. GTF: Basement of Mansfield House, Votey parking lot, 656-3349(O) Office Hours: CLASS ROOM & CLASS HOURS: 5:00 pm - 8:45 pm TR, LAFAYE HALL L411 May 20, 2013 - June 28, 2013 TEXTBOOK: Applied Linear Regression Models by Kutner, Nachtsheim and Neter, 4th Ed., Irwin. ISBN: 0-07-238691-6 JMP, SAS/ETS --- recommended. COURSE OUTLINE: Regression analysis is a powerful, basic statistical technique which is being used in many fields. All statistical computing systems provide linear regression programs. The insight necessary to answer important questions such as - How can we find out if the regression assumptions are wrong? If they are, what do we do? Why is the order of entry of terms into the model important? Should we transform the response variables? If so, how? Should terms be dropped from a polynomial model? How can different batches of raw material be introduced in the regression equation? Are individual confidence intervals for regression coefficients good or bad? Why are coefficients tested with both t and F statistics? What is a general linear models? When should you use stepwise regression to build a model? What is nonlinear estimation? To obtain these goals, we will cover Chaps. 1 - 3, 5, Chaps. 6 - 11, and Chap. 14. Some sections will be skipped. Chap. 1 -- Linear Regression with One Predictor Variable Chap. 2 -- Inferences in Regression and Correlation Analysis Chap. 3 -- Diagnostics and Remedial Measures Chap. 5 -- Matrix approach to Simple Linear Regression Analysis Chap. 6 -- Multiple Regression I Chap. 7 -- Multiple Regression II Chap. 8 -- Regression Models for Quantitative and Qualitative Predictors Chap. 9 -- Building the Regression Model I: Model Selection and Validation Chap. 10 -- Building the Regression Model II: Diagnostics Chap. 11 -- Building the Regression Model III: Remedial Measures Chap. 14 -- Logistic Regression, Poisson Regression, and Generalized Linear Models Some sections will be skipped. GRADING: HW/Project - 10% Exam I - 40% (Tue., 6/11/13) Final - 50% (Thu., 6/27/13, Comprehensive) All students are expected to regulate their class attendance.