Regression Analysis

Course Description

Simple linear regression including the method of least squares, statistical inference about regression coefficients, methods of measuring model adequacy, prediction, multiple linear regression including estimation of parameters, using matrix in regression, confidence intervals, tests of hypotheses and prediction, model adequacy checking and multicollinearity, residual analysis, polynomial regression, variable selection and model building, nonlinear regression, applications using computer.

Course Objectives & Outcomes

The objectives of this course are :

• Studying the statistical methods used in regression analysis.
•  Acquaint students with least square methods and concept of linear regression and its applications.
•  Develop the ability to build regression models.
•  Gain familiarity with use of modern statistical software packages for building a statistical model.
•  Estimate the parameters of a linear regression models and evaluate its adequacy.

Upon successful completion of this course, the student will be able to:

•  Understand method and concept of simple and multiple regression and understand the basic idea and the assumptions of the least squares method.
•  Write simple and multiple linear regression models in matrix format.
• Build regression models.
• Present the results using available statistical software.

References
1. John Fox (2015) , Analysis & Generalized Linear Models, 3rd edition, SAGE.
ISBN-13: 978-0761930426 ISBN-10: 0761930426 .
2. John Fox and Harvey Sanford Weisberg, (2011), An R Companion to Applied Regression, 2nd edition, SAGE, ISBN-13:978-0761930426.
3. Sanford Weisberg, (2005), Applied linear regression, 3rd edition, John Wiley & Sons, Inc.
ISBN 0-471-66379-4 .

Course ID: STAT 410