Elements of Multiple Regression Analysis: Two Independent Variables Yong Sept. 2010.

Slides:



Advertisements
Similar presentations
11-1 Empirical Models Many problems in engineering and science involve exploring the relationships between two or more variables. Regression analysis.
Advertisements

Correlation and Linear Regression.
6-1 Introduction To Empirical Models 6-1 Introduction To Empirical Models.
Regression Analysis Module 3. Regression Regression is the attempt to explain the variation in a dependent variable using the variation in independent.
Regression Analysis Once a linear relationship is defined, the independent variable can be used to forecast the dependent variable. Y ^ = bo + bX bo is.
Multiple Regression. Outline Purpose and logic : page 3 Purpose and logic : page 3 Parameters estimation : page 9 Parameters estimation : page 9 R-square.
Describing Relationships Using Correlation and Regression
Education 793 Class Notes Joint Distributions and Correlation 1 October 2003.
Regression single and multiple. Overview Defined: A model for predicting one variable from other variable(s). Variables:IV(s) is continuous, DV is continuous.
© McGraw-Hill Higher Education. All Rights Reserved. Chapter 2F Statistical Tools in Evaluation.
LINEAR REGRESSION: Evaluating Regression Models Overview Assumptions for Linear Regression Evaluating a Regression Model.
LINEAR REGRESSION: Evaluating Regression Models. Overview Assumptions for Linear Regression Evaluating a Regression Model.
Chapter 10 Simple Regression.
Multiple Regression Involves the use of more than one independent variable. Multivariate analysis involves more than one dependent variable - OMS 633 Adding.
Topics: Regression Simple Linear Regression: one dependent variable and one independent variable Multiple Regression: one dependent variable and two or.
Ch. 14: The Multiple Regression Model building
11-1 Empirical Models Many problems in engineering and science involve exploring the relationships between two or more variables. Regression analysis.
Multiple Regression Research Methods and Statistics.
Correlation and Regression Analysis
Multiple Regression Dr. Andy Field.
Multiple Regression Models
SPSS Statistical Package for Social Sciences Multiple Regression Department of Psychology California State University Northridge
Relationships Among Variables
Review for Final Exam Some important themes from Chapters 9-11 Final exam covers these chapters, but implicitly tests the entire course, because we use.
Review Guess the correlation. A.-2.0 B.-0.9 C.-0.1 D.0.1 E.0.9.
Correlation & Regression
Multiple Linear Regression Response Variable: Y Explanatory Variables: X 1,...,X k Model (Extension of Simple Regression): E(Y) =  +  1 X 1 +  +  k.
Example of Simple and Multiple Regression
Lecture 15 Basics of Regression Analysis
Marketing Research Aaker, Kumar, Day and Leone Tenth Edition
Introduction to Linear Regression and Correlation Analysis
Wednesday PM  Presentation of AM results  Multiple linear regression Simultaneous Simultaneous Stepwise Stepwise Hierarchical Hierarchical  Logistic.
Statistics for Managers Using Microsoft Excel, 4e © 2004 Prentice-Hall, Inc. Chap 12-1 Chapter 12 Simple Linear Regression Statistics for Managers Using.
Chapter 13: Inference in Regression
Copyright © 2013, 2010 and 2007 Pearson Education, Inc. Chapter Inference on the Least-Squares Regression Model and Multiple Regression 14.
Copyright © 2010 Pearson Education, Inc Chapter Seventeen Correlation and Regression.
CHAPTER 14 MULTIPLE REGRESSION
Chapter 12 Examining Relationships in Quantitative Research Copyright © 2013 by The McGraw-Hill Companies, Inc. All rights reserved. McGraw-Hill/Irwin.
Statistics for the Social Sciences Psychology 340 Fall 2013 Correlation and Regression.
Multiple Linear Regression. Purpose To analyze the relationship between a single dependent variable and several independent variables.
Part IV Significantly Different Using Inferential Statistics Chapter 15 Using Linear Regression Predicting Who’ll Win the Super Bowl.
Part IV Significantly Different: Using Inferential Statistics
Chapter 7 Relationships Among Variables What Correlational Research Investigates Understanding the Nature of Correlation Positive Correlation Negative.
Regression Chapter 16. Regression >Builds on Correlation >The difference is a question of prediction versus relation Regression predicts, correlation.
SW388R6 Data Analysis and Computers I Slide 1 Multiple Regression Key Points about Multiple Regression Sample Homework Problem Solving the Problem with.
Regression Lesson 11. The General Linear Model n Relationship b/n predictor & outcome variables form straight line l Correlation, regression, t-tests,
© 2006 by The McGraw-Hill Companies, Inc. All rights reserved. 1 Chapter 12 Testing for Relationships Tests of linear relationships –Correlation 2 continuous.
 Relationship between education level, income, and length of time out of school  Our new regression equation: is the predicted value of the dependent.
Chapter 22: Building Multiple Regression Models Generalization of univariate linear regression models. One unit of data with a value of dependent variable.
Copyright © 2010 Pearson Education, Inc Chapter Seventeen Correlation and Regression.
Research Methodology Lecture No :26 (Hypothesis Testing – Relationship)
Applied Quantitative Analysis and Practices LECTURE#28 By Dr. Osman Sadiq Paracha.
Chapter 8 Relationships Among Variables. Outline What correlational research investigates Understanding the nature of correlation What the coefficient.
Week of March 23 Partial correlations Semipartial correlations
Chapter 11 REGRESSION Multiple Regression  Uses  Explanation  Prediction.
Regression. Why Regression? Everything we’ve done in this class has been regression: When you have categorical IVs and continuous DVs, the ANOVA framework.
Chapter 12 REGRESSION DIAGNOSTICS AND CANONICAL CORRELATION.
Topics: Multiple Regression Analysis (MRA)
Multiple Regression.
Simple Bivariate Regression
Correlation, Bivariate Regression, and Multiple Regression
Multiple Regression.
Regression.
بحث في التحليل الاحصائي SPSS بعنوان :
Multiple Regression – Part II
Multiple Regression.
Simple Linear Regression
Simple Linear Regression
Regression Analysis.
Regression Part II.
Presentation transcript:

Elements of Multiple Regression Analysis: Two Independent Variables Yong Sept. 2010

Why Multiple Regression?  In real world, using only one predictor (IV) to interpret or predict a outcome variable (DV) is rare. Mostly, we need several IV’s.  Multiple regression (Pearson, 1908) is to investigate the relationship between several independent or predictor variables and a dependent or criterion variable.

The prediction equation in multiple regression Y’ = predicted Y score a = intercept b 1 … b k = regression coefficients X 1 … X k = scores of IVs With two IV’s:

Calculation of basic statistics 1  Calculation with two IV’s is similar to one IV. However, it is not hard but tedious.  We need knowledge of matrix operations to perform calculations with 3 or more IV’s.  Good news is that we can have the computer do the calculations!

Calculation of basic statistics 2

Calculation of basic statistics 3

Why calculations, as always?  Intercept (a) & regression coefficients (b’s) !

Brain exercise  Now, we have the regression line!  What’s next?  The predicted Y or Y’!  Then what?  Deviation due to regression ( ) and the regression sum of squares ( ).  Deviation due to residuals ( ) and the residual sum of squares ( ).

Sum of squares  Recall that we have plenty ways to calculate the sum of squares. Some methods allow us to calculate sum of squares without using Y’:  Remember, we need Y’ to calculate residuals, which are essential for regression diagnostics (chapter 3).

Squared multiple correlation coefficient  R-square indicates the proportion of variance of the DV (Y) accounted for by the IV’s (X’s).  Note that R 2 is equivalent to for two IV’s.

Test of significance of R 2 F test: if R 2 is significantly different from 0. Rule of thumb: We reject H 0 when the calculated F is greater than the table (critical) value or the calculated probability is less than α.  significance level fail to reject H 0 reject H 0 F critical Probability, p

Test of significance of individual b’s  T-test (mostly two-tailed, except that we can rule out one direction): if b is significantly different from 0. Rule of thumb: We reject H 0 when the absolute value of calculated T is greater than the table (critical) value or the calculated probability is less than α.  fail to reject H 0 reject H 0 

Test of R 2 vs. test of b  Test of R 2 is equivalent to testing all the b’s simultaneously.  Test of a given b for significance is to determine whether it differs from 0 while controlling for the effects of the other IV’s.  For simple linear regression, they are equivalent ( ).

Confidence interval  Definition: If an experiment was repeated many times, 100(1-α)% of these intervals would contain µ.  If the CI does not include 0, we reject H 0 and conclude that the given regression coefficient significantly differs from 0.

Test of increments in proportion of variance accounted for (R 2 change)  In multiple linear regression, we could test amount of R 2 increases or decreases when a given IV or a set of variables are added to or deleted from the regression equation.

Test of increments in proportion of variance accounted for (R 2 change)  The test is equivalent to testing significance of individual b if one IV is added to or deleted from the regression equation.  Note that the R 2 change caused by a given IV or a set of IV’s depends on the order of addition or deletion.

Commonly used methods of adding or deleting variables  Enter: enter all IV’s at once in a single model  Stepwise: enter IV’s one by one in several models commonly based on R 2  Forward: enter IV’s one by one based on strength of correlation with DV.  Backward: enter all IV’s and delete weakest one unless it significantly affects the model.  Hierarchical: enter IV’s (one or more at a time) according to certain theoretical framework.

Standardized regression coefficient (β, beta)  In SPSS (now PASW) output, we have something like this:  Is it a population parameter?

Standardized regression coefficient (β, beta)  Sample unstandardized regression coefficient (b) is the expected change in Y associated with one measurement unit change of in X.  Sample standardized regression coefficient (β) is the expected change in standard deviation of Y associated with a change of one standard deviation in X.

Standardized regression coefficient (β, beta)  The regression equation now is:  Note that the α disappears because standardized score for a constant is always 0.  β could be used to determine the relative contribution of individual IV to account for variance in DV.

What about the correlation coefficients (r’s)?  Later, we will discuss the correlation coefficients in details, mostly in chapter 7 (Statistical Control: Partial and Semipartial Correlation).

Remarks  Multiple regression is an upgraded version of simple linear regression and its interpretation is similar to simple linear regression.  We need emphasize on contributions of each individual IV’s.  To some extent, multiple IV’s have better explanation and prediction on the DV – it is not always true.