STAT 533
Regression Analysis
Semester I, 1428/1429 A.H.

Instructor: Habib A. Ismail

Office: 2B16 building # 4

Time: Sunday  12:30 - 3:00 PM

 Text Book:

Title: Applied Linear Regression Models


John Neter, William Wasserman, and Michael H. Kutner

Edition: Second Edition

 Grading Policy:

First mid-term test : 20%

Second mid-term test : 20%

Final Test : 60%

Homework and Testing Policies:

University and College policies on homeworks and

examinations will be strictly enforced.

Students are to work independently on tests.

Tests are to be submitted on time.

 Course Outline: 

Course number and symbol

STAT 535

Course name

Regression Analysis

Credit units



Stat 335






College of science

Statistics major





Course definition and description (in brief)

Multiple linear regression - Residual analysis - Polynomial regression - Indicator variables - Model building and variable selection - Influential data - Ridge regression - Nonlinear regression.

Main topics

Detailed topics enclosed.


The aim of this course is to train and teach students practically and theoretically the bases and principals of regression analysis in order to be able to exercise the contents of this field in their actual life.

Teaching method

Lectures and Exercises.


Title: Applied Linear Regression Models

Authors: John Neter, William Wasserman, and Michael H. Kutner

Publisher: Richard D. Irwin, Inc.            Year: 1983

Main references

1- Regression Analysis: Concepts and Applications by Franklin A. Graybill, and Hariharan K. Iyer

2- Applied Linear Regression by Sanford Weisberg


Evaluation method

1- Two mid-term tests

2- Exercises, assignments and projects

3- Final examination

First mid-term test

Two-hours test

Sixth or seventh week

Second mid-term test

Two-hours test

Eleventh or twelfth week

Grades allocation

Mid-term 40 marks

Final examination 60 marks

Duration of finals

Three hours

Date approved


Main topics (Detailed topics )

Multiple linear regression

Multiple regression models, General linear regression model in matrix terms, Least squares estimators, Analysis of variance results, Inferences about regression parameters, Inferences about mean response, Predictions of new observations, Decomposition of SSR into extra sums of squares,  Coefficients of partial determination, Testing hypotheses concerning regression coefficients in, Matrix formulation of general linear test,

Residual analysis

Residuals, Graphic analysis of residuals, F test for lack of fit, Remedial measures,

Polynomial regression

Polynomial regression models, Estimating the maximum or minimum of a quadratic regression function,  Some further comments on polynomial regression.

Indicator variables

One independent qualitative variable, Model containing interaction effects, More complex models, Other uses of independent indicator variables. Some considerations in using independent indicator variables. Dependent indicator variable. Linear regression with dependent indicator variable, Logistic response function.

Model building and variable selection

Selection of independent variables,  Nature of the problem,  All possible regression models,  Stepwise regression,  Implementation of selection procedures.

Influential data

Multicollinearity, influential observations, and other topics in regression analysis,  Reparameterization to improve computational accuracy, Problems of multicollinearity, Variance inflation factors and other methods of detecting


Ridge regression

Ridge regression and other remedial measures for multicollinearity, Identification of outlying observations, Identification of influential observations and remedial measures.

Nonlinear regression

Linear, intrinsically linear, and nonlinear regression models, Least squares estimation in nonlinear regression, Inferences about nonlinear regression parameters.