SlideShare ist ein Scribd-Unternehmen logo
1 von 12
Downloaden Sie, um offline zu lesen
Chapter 15

 ISDS 2001
  Matt Levy
Multiple Regression Model and Equation
What we learned in SLR, is also applicable in Multiple
Regression.

The multiple regression model simply extends SLR to include
more than 1 independent variable.

Hence we augment our simple linear model accomodate this:

y = β0 + β1x1 + β2x2 + ... + βpxp + ε

Additionally, since we still assume the expected value of ε to be
zero, we show the multiple regression equation as follows:

E(y) = β0 + β1x1 + β2x2 + ... + βpxp
Estimated Multiple Regression Equation

If β0, β1,...βp were known the equation on the previous slide
could be used to compute the mean value of y at given values
of x1, x2, ..., xp.

But we don't know them, so we need estimates b0, b1, ..., bp

Thus we arrive at the Estimated Regression Equation:

ŷ = b0 + b1x1 + b2x2 + ... + bpxp
The Estimation Process for Multiple Regression
Least Squares Method
To estimate our beta's, the objective is the same as SLR.

That is we seek to minimize the difference between our actual
dependent variable (y) and the prediction for that dependent
variable (ŷ).

Least Square Criterion: min Σ(yi - ŷi)2

In SLR, we had a relatively easy way to obtain our estimates.

In multiple regression, this is not so easy:
   B = (X'X)-1X'Y

So we rely on statistical computing packages to do this for us.
Interpretation of Coefficients (β) in
Multiple Regression
Now that we have > 1 independent variables, we must be aware of the consequences of adding
multiple independent variables.

Notice from the example in Ch.15, that a b1 estimate computed with 1 independent variable (SLR)
will NOT be the same when additional independent variables are added.

In SLR we interpreted b1 as an estimate of the change in y for a 1 unit change in x.

In multiple regression, bi is an estimate of the change in y for a 1 unit change in xi when all other
independent variables are held constant. (For example, when they are all 0)

Take note also that now we can easily throw in as many independent variables as we want.

This will increase our explained variance, and our R2...So this is good, right??? Wrong.

While this may increase our ability to predict, it will also make our model increasingly complex.

Statistical power is achieved through accurate prediction with least amount of variables.

In the coming sections we will look at additional measures for 'model parsimony'...that is models
that 'do the most with the least'.
Model Assumptions
Our assumptions in multiple regression parallel those in
SLR. For emphasis, lets briefly review. (Also look at 15.11)

1. E(ε) = 0; Therefore E(y) = β0 + β1x1 + β2x2 + ... + βpxp
2. Var(ε) = σ2 and is the same for all values of x; Therefore the
   variance about the regression line also equals σ2 and is the
   same for all values of x.
3. The values of ε are independent; Therefore the values of ε
   for any set of x values are not related to any other set of x
   values.
4. ε is normally distributed random variable; Therefore y is
   normally distributed.
Testing for Significance
In multiple regression significance testing carries slightly
different meaning than in SLR.

1. F-Test: Tests for a significant relationship between the
dependent variable and the set of all independent
variables. We refer to this as the test for overall significance.

2. If the F-Test shows overall significance, than we use the t-
test to check the significance for each of the independent
variables. We conduct the t-test on each of the independent
variables. We refer to the t-test as the test for individual
significance.
F-Test
In multiple regression, we test that none of the parameters are
equal to zero:

  H0: β0 = β1 = ... = βp = 0
  Ha: One or more of the parameters are equal to zero.

Remember that F = MSR/MSE.

And in multiple regression:

MSR = SSR/p
MSE = SSE/(n - p - 1)

And we reject H0 if our p-value < α
T-Test
Remember we test for each parameter.

For any βi:

   H0: βi = 0
   Ha: βi ≠ 0

t = βi/sbi

And we reject H0 if our p-value < α
Multicollinearity

This is essentially the correlation among independent variables.

We care about this because we want our independent variables to measure
significantly different things when predicting our dependent variable.

While in practice there is always some multicollinearity, we need to try and
eliminate as much as we can.

A simple test of multicollinearity is with the sample correlation (rx1x2) for any
two independent variables.

If the sample correlation exceeds .7 for any two independent variables we must
take measures to reduce multicollinearity, for example, removing one of the two
highly correlated variables from the model.
The End

Thats it for Ch. 15.

Hope you have recovered from Mardi Gras next time I see you!

Weitere ähnliche Inhalte

Was ist angesagt?

Error analysis in numerical integration
Error analysis in numerical integrationError analysis in numerical integration
Error analysis in numerical integrationAmenahGondal1
 
25 the ratio, root, and ratio comparison test x
25 the ratio, root, and ratio  comparison test x25 the ratio, root, and ratio  comparison test x
25 the ratio, root, and ratio comparison test xmath266
 
Artifact 3 clemson
Artifact 3 clemsonArtifact 3 clemson
Artifact 3 clemsonclemsonj11
 
Solving stepwise regression problems
Solving stepwise regression problemsSolving stepwise regression problems
Solving stepwise regression problemsSoma Sinha Roy
 
Exponential formula presentation
Exponential formula presentationExponential formula presentation
Exponential formula presentationOliver Zhang
 
Regression Analysis presentation by Al Arizmendez and Cathryn Lottier
Regression Analysis presentation by Al Arizmendez and Cathryn LottierRegression Analysis presentation by Al Arizmendez and Cathryn Lottier
Regression Analysis presentation by Al Arizmendez and Cathryn LottierAl Arizmendez
 
Regression analysis ppt
Regression analysis pptRegression analysis ppt
Regression analysis pptElkana Rorio
 
Visual Explanation of Ridge Regression and LASSO
Visual Explanation of Ridge Regression and LASSOVisual Explanation of Ridge Regression and LASSO
Visual Explanation of Ridge Regression and LASSOKazuki Yoshida
 
Changing the subject of a formula (roots and powers)
Changing the subject of a formula (roots and powers)Changing the subject of a formula (roots and powers)
Changing the subject of a formula (roots and powers)Alona Hall
 
Exact &amp; non exact differential equation
Exact &amp; non exact differential equationExact &amp; non exact differential equation
Exact &amp; non exact differential equationAlaminMdBabu
 
G6 m3-b-lesson 11-t
G6 m3-b-lesson 11-tG6 m3-b-lesson 11-t
G6 m3-b-lesson 11-tmlabuski
 
Presentation on regression analysis
Presentation on regression analysisPresentation on regression analysis
Presentation on regression analysisSujeet Singh
 
Simple linear regression
Simple linear regressionSimple linear regression
Simple linear regressionpankaj8108
 
G6 m3-b-lesson 11-s
G6 m3-b-lesson 11-sG6 m3-b-lesson 11-s
G6 m3-b-lesson 11-smlabuski
 

Was ist angesagt? (20)

Error analysis in numerical integration
Error analysis in numerical integrationError analysis in numerical integration
Error analysis in numerical integration
 
25 the ratio, root, and ratio comparison test x
25 the ratio, root, and ratio  comparison test x25 the ratio, root, and ratio  comparison test x
25 the ratio, root, and ratio comparison test x
 
Artifact 3 clemson
Artifact 3 clemsonArtifact 3 clemson
Artifact 3 clemson
 
Regression
RegressionRegression
Regression
 
Solving stepwise regression problems
Solving stepwise regression problemsSolving stepwise regression problems
Solving stepwise regression problems
 
F-Compact operator1_p1-2
F-Compact operator1_p1-2F-Compact operator1_p1-2
F-Compact operator1_p1-2
 
Simple linear regression
Simple linear regressionSimple linear regression
Simple linear regression
 
Exponential formula presentation
Exponential formula presentationExponential formula presentation
Exponential formula presentation
 
Regression Analysis presentation by Al Arizmendez and Cathryn Lottier
Regression Analysis presentation by Al Arizmendez and Cathryn LottierRegression Analysis presentation by Al Arizmendez and Cathryn Lottier
Regression Analysis presentation by Al Arizmendez and Cathryn Lottier
 
Regression analysis ppt
Regression analysis pptRegression analysis ppt
Regression analysis ppt
 
3 handouts section3-11
3 handouts section3-113 handouts section3-11
3 handouts section3-11
 
Visual Explanation of Ridge Regression and LASSO
Visual Explanation of Ridge Regression and LASSOVisual Explanation of Ridge Regression and LASSO
Visual Explanation of Ridge Regression and LASSO
 
Algebra review
Algebra reviewAlgebra review
Algebra review
 
Ridge regression
Ridge regressionRidge regression
Ridge regression
 
Changing the subject of a formula (roots and powers)
Changing the subject of a formula (roots and powers)Changing the subject of a formula (roots and powers)
Changing the subject of a formula (roots and powers)
 
Exact &amp; non exact differential equation
Exact &amp; non exact differential equationExact &amp; non exact differential equation
Exact &amp; non exact differential equation
 
G6 m3-b-lesson 11-t
G6 m3-b-lesson 11-tG6 m3-b-lesson 11-t
G6 m3-b-lesson 11-t
 
Presentation on regression analysis
Presentation on regression analysisPresentation on regression analysis
Presentation on regression analysis
 
Simple linear regression
Simple linear regressionSimple linear regression
Simple linear regression
 
G6 m3-b-lesson 11-s
G6 m3-b-lesson 11-sG6 m3-b-lesson 11-s
G6 m3-b-lesson 11-s
 

Andere mochten auch

Wireless Technology
Wireless TechnologyWireless Technology
Wireless Technologyrromro
 
Present, Past and Future of the Telephony in Mexico
Present, Past and Future of the Telephony in MexicoPresent, Past and Future of the Telephony in Mexico
Present, Past and Future of the Telephony in Mexicorromro
 
File storageandbackupsystems
File storageandbackupsystemsFile storageandbackupsystems
File storageandbackupsystemsMicheline LeBlanc
 
Story Board Consejo South Park
Story Board Consejo South ParkStory Board Consejo South Park
Story Board Consejo South Parkrromro
 
Presentation Week 2
Presentation Week 2Presentation Week 2
Presentation Week 2rromro
 
Uwtwtrbook Tweeting In Latin America
Uwtwtrbook Tweeting In Latin AmericaUwtwtrbook Tweeting In Latin America
Uwtwtrbook Tweeting In Latin Americarromro
 
MAGRAY- A MARTIAL KASHMIRI TRIBE/CASTE/RACE
MAGRAY- A MARTIAL KASHMIRI TRIBE/CASTE/RACEMAGRAY- A MARTIAL KASHMIRI TRIBE/CASTE/RACE
MAGRAY- A MARTIAL KASHMIRI TRIBE/CASTE/RACEguest831335
 
Create Tutorials and Videos Using SMART Recorder
Create Tutorials and Videos Using SMART RecorderCreate Tutorials and Videos Using SMART Recorder
Create Tutorials and Videos Using SMART Recordermr_tripp
 
Chapter 20 Lecture Notes
Chapter 20 Lecture NotesChapter 20 Lecture Notes
Chapter 20 Lecture NotesMatthew L Levy
 
Loren Gray's HSMAI ROC 2016 Presentation
Loren Gray's HSMAI ROC 2016 PresentationLoren Gray's HSMAI ROC 2016 Presentation
Loren Gray's HSMAI ROC 2016 PresentationLoren Gray, CHDM
 

Andere mochten auch (18)

Chapter 4 R Part I
Chapter 4 R Part IChapter 4 R Part I
Chapter 4 R Part I
 
Chapter 14 Part Ii
Chapter 14 Part IiChapter 14 Part Ii
Chapter 14 Part Ii
 
Chapter 18 Part I
Chapter 18 Part IChapter 18 Part I
Chapter 18 Part I
 
Wireless Technology
Wireless TechnologyWireless Technology
Wireless Technology
 
Present, Past and Future of the Telephony in Mexico
Present, Past and Future of the Telephony in MexicoPresent, Past and Future of the Telephony in Mexico
Present, Past and Future of the Telephony in Mexico
 
Chapter 14R
Chapter 14RChapter 14R
Chapter 14R
 
File storageandbackupsystems
File storageandbackupsystemsFile storageandbackupsystems
File storageandbackupsystems
 
Story Board Consejo South Park
Story Board Consejo South ParkStory Board Consejo South Park
Story Board Consejo South Park
 
Chapter 5R
Chapter 5RChapter 5R
Chapter 5R
 
Chapter 15R Lecture
Chapter 15R LectureChapter 15R Lecture
Chapter 15R Lecture
 
Presentation Week 2
Presentation Week 2Presentation Week 2
Presentation Week 2
 
Uwtwtrbook Tweeting In Latin America
Uwtwtrbook Tweeting In Latin AmericaUwtwtrbook Tweeting In Latin America
Uwtwtrbook Tweeting In Latin America
 
Chapter 14 Part I
Chapter 14 Part IChapter 14 Part I
Chapter 14 Part I
 
MAGRAY- A MARTIAL KASHMIRI TRIBE/CASTE/RACE
MAGRAY- A MARTIAL KASHMIRI TRIBE/CASTE/RACEMAGRAY- A MARTIAL KASHMIRI TRIBE/CASTE/RACE
MAGRAY- A MARTIAL KASHMIRI TRIBE/CASTE/RACE
 
Create Tutorials and Videos Using SMART Recorder
Create Tutorials and Videos Using SMART RecorderCreate Tutorials and Videos Using SMART Recorder
Create Tutorials and Videos Using SMART Recorder
 
Chapter 4R Part II
Chapter 4R Part IIChapter 4R Part II
Chapter 4R Part II
 
Chapter 20 Lecture Notes
Chapter 20 Lecture NotesChapter 20 Lecture Notes
Chapter 20 Lecture Notes
 
Loren Gray's HSMAI ROC 2016 Presentation
Loren Gray's HSMAI ROC 2016 PresentationLoren Gray's HSMAI ROC 2016 Presentation
Loren Gray's HSMAI ROC 2016 Presentation
 

Ähnlich wie Chapter 15

For this assignment, use the aschooltest.sav dataset.The d
For this assignment, use the aschooltest.sav dataset.The dFor this assignment, use the aschooltest.sav dataset.The d
For this assignment, use the aschooltest.sav dataset.The dMerrileeDelvalle969
 
Two-Variable (Bivariate) RegressionIn the last unit, we covered
Two-Variable (Bivariate) RegressionIn the last unit, we covered Two-Variable (Bivariate) RegressionIn the last unit, we covered
Two-Variable (Bivariate) RegressionIn the last unit, we covered LacieKlineeb
 
Get Multiple Regression Assignment Help
Get Multiple Regression Assignment Help Get Multiple Regression Assignment Help
Get Multiple Regression Assignment Help HelpWithAssignment.com
 
Chapter 15Multiple Regression and Model BuildingCo
Chapter 15Multiple Regression and Model BuildingCoChapter 15Multiple Regression and Model BuildingCo
Chapter 15Multiple Regression and Model BuildingCoEstelaJeffery653
 
Detail Study of the concept of Regression model.pptx
Detail Study of the concept of  Regression model.pptxDetail Study of the concept of  Regression model.pptx
Detail Study of the concept of Regression model.pptxtruptikulkarni2066
 
Data Science - Part XII - Ridge Regression, LASSO, and Elastic Nets
Data Science - Part XII - Ridge Regression, LASSO, and Elastic NetsData Science - Part XII - Ridge Regression, LASSO, and Elastic Nets
Data Science - Part XII - Ridge Regression, LASSO, and Elastic NetsDerek Kane
 
Logistic regression
Logistic regressionLogistic regression
Logistic regressionRupak Roy
 
Chapter6
Chapter6Chapter6
Chapter6Vu Vo
 
Linear Regression
Linear Regression Linear Regression
Linear Regression Rupak Roy
 
The linear regression model: Theory and Application
The linear regression model: Theory and ApplicationThe linear regression model: Theory and Application
The linear regression model: Theory and ApplicationUniversity of Salerno
 
Unit-4-BSR-11-1-2024 (1)linear programming.pptx
Unit-4-BSR-11-1-2024 (1)linear programming.pptxUnit-4-BSR-11-1-2024 (1)linear programming.pptx
Unit-4-BSR-11-1-2024 (1)linear programming.pptxIronMan665214
 
IBM401 Lecture 5
IBM401 Lecture 5IBM401 Lecture 5
IBM401 Lecture 5saark
 
SM_d89ccf05-7de1-4a30-a134-3143e9b3bf3f_38.pptx
SM_d89ccf05-7de1-4a30-a134-3143e9b3bf3f_38.pptxSM_d89ccf05-7de1-4a30-a134-3143e9b3bf3f_38.pptx
SM_d89ccf05-7de1-4a30-a134-3143e9b3bf3f_38.pptxManjulasingh17
 
Economics Assignment Sample Problems Set 6 with Solutions
Economics Assignment Sample Problems Set 6 with SolutionsEconomics Assignment Sample Problems Set 6 with Solutions
Economics Assignment Sample Problems Set 6 with SolutionsHebrew Johnson
 
Non linearregression 4+
Non linearregression 4+Non linearregression 4+
Non linearregression 4+Ricardo Solano
 

Ähnlich wie Chapter 15 (20)

For this assignment, use the aschooltest.sav dataset.The d
For this assignment, use the aschooltest.sav dataset.The dFor this assignment, use the aschooltest.sav dataset.The d
For this assignment, use the aschooltest.sav dataset.The d
 
Two-Variable (Bivariate) RegressionIn the last unit, we covered
Two-Variable (Bivariate) RegressionIn the last unit, we covered Two-Variable (Bivariate) RegressionIn the last unit, we covered
Two-Variable (Bivariate) RegressionIn the last unit, we covered
 
Get Multiple Regression Assignment Help
Get Multiple Regression Assignment Help Get Multiple Regression Assignment Help
Get Multiple Regression Assignment Help
 
Chapter 15Multiple Regression and Model BuildingCo
Chapter 15Multiple Regression and Model BuildingCoChapter 15Multiple Regression and Model BuildingCo
Chapter 15Multiple Regression and Model BuildingCo
 
Detail Study of the concept of Regression model.pptx
Detail Study of the concept of  Regression model.pptxDetail Study of the concept of  Regression model.pptx
Detail Study of the concept of Regression model.pptx
 
Data Science - Part XII - Ridge Regression, LASSO, and Elastic Nets
Data Science - Part XII - Ridge Regression, LASSO, and Elastic NetsData Science - Part XII - Ridge Regression, LASSO, and Elastic Nets
Data Science - Part XII - Ridge Regression, LASSO, and Elastic Nets
 
Logistic regression
Logistic regressionLogistic regression
Logistic regression
 
Chapter6
Chapter6Chapter6
Chapter6
 
Linear Regression
Linear Regression Linear Regression
Linear Regression
 
Logistics regression
Logistics regressionLogistics regression
Logistics regression
 
Chapter14
Chapter14Chapter14
Chapter14
 
The linear regression model: Theory and Application
The linear regression model: Theory and ApplicationThe linear regression model: Theory and Application
The linear regression model: Theory and Application
 
Unit-4-BSR-11-1-2024 (1)linear programming.pptx
Unit-4-BSR-11-1-2024 (1)linear programming.pptxUnit-4-BSR-11-1-2024 (1)linear programming.pptx
Unit-4-BSR-11-1-2024 (1)linear programming.pptx
 
Chapter14
Chapter14Chapter14
Chapter14
 
Corr And Regress
Corr And RegressCorr And Regress
Corr And Regress
 
IBM401 Lecture 5
IBM401 Lecture 5IBM401 Lecture 5
IBM401 Lecture 5
 
Talk 4
Talk 4Talk 4
Talk 4
 
SM_d89ccf05-7de1-4a30-a134-3143e9b3bf3f_38.pptx
SM_d89ccf05-7de1-4a30-a134-3143e9b3bf3f_38.pptxSM_d89ccf05-7de1-4a30-a134-3143e9b3bf3f_38.pptx
SM_d89ccf05-7de1-4a30-a134-3143e9b3bf3f_38.pptx
 
Economics Assignment Sample Problems Set 6 with Solutions
Economics Assignment Sample Problems Set 6 with SolutionsEconomics Assignment Sample Problems Set 6 with Solutions
Economics Assignment Sample Problems Set 6 with Solutions
 
Non linearregression 4+
Non linearregression 4+Non linearregression 4+
Non linearregression 4+
 

Chapter 15

  • 1. Chapter 15 ISDS 2001 Matt Levy
  • 2. Multiple Regression Model and Equation What we learned in SLR, is also applicable in Multiple Regression. The multiple regression model simply extends SLR to include more than 1 independent variable. Hence we augment our simple linear model accomodate this: y = β0 + β1x1 + β2x2 + ... + βpxp + ε Additionally, since we still assume the expected value of ε to be zero, we show the multiple regression equation as follows: E(y) = β0 + β1x1 + β2x2 + ... + βpxp
  • 3. Estimated Multiple Regression Equation If β0, β1,...βp were known the equation on the previous slide could be used to compute the mean value of y at given values of x1, x2, ..., xp. But we don't know them, so we need estimates b0, b1, ..., bp Thus we arrive at the Estimated Regression Equation: ŷ = b0 + b1x1 + b2x2 + ... + bpxp
  • 4. The Estimation Process for Multiple Regression
  • 5. Least Squares Method To estimate our beta's, the objective is the same as SLR. That is we seek to minimize the difference between our actual dependent variable (y) and the prediction for that dependent variable (ŷ). Least Square Criterion: min Σ(yi - ŷi)2 In SLR, we had a relatively easy way to obtain our estimates. In multiple regression, this is not so easy: B = (X'X)-1X'Y So we rely on statistical computing packages to do this for us.
  • 6. Interpretation of Coefficients (β) in Multiple Regression Now that we have > 1 independent variables, we must be aware of the consequences of adding multiple independent variables. Notice from the example in Ch.15, that a b1 estimate computed with 1 independent variable (SLR) will NOT be the same when additional independent variables are added. In SLR we interpreted b1 as an estimate of the change in y for a 1 unit change in x. In multiple regression, bi is an estimate of the change in y for a 1 unit change in xi when all other independent variables are held constant. (For example, when they are all 0) Take note also that now we can easily throw in as many independent variables as we want. This will increase our explained variance, and our R2...So this is good, right??? Wrong. While this may increase our ability to predict, it will also make our model increasingly complex. Statistical power is achieved through accurate prediction with least amount of variables. In the coming sections we will look at additional measures for 'model parsimony'...that is models that 'do the most with the least'.
  • 7. Model Assumptions Our assumptions in multiple regression parallel those in SLR. For emphasis, lets briefly review. (Also look at 15.11) 1. E(ε) = 0; Therefore E(y) = β0 + β1x1 + β2x2 + ... + βpxp 2. Var(ε) = σ2 and is the same for all values of x; Therefore the variance about the regression line also equals σ2 and is the same for all values of x. 3. The values of ε are independent; Therefore the values of ε for any set of x values are not related to any other set of x values. 4. ε is normally distributed random variable; Therefore y is normally distributed.
  • 8. Testing for Significance In multiple regression significance testing carries slightly different meaning than in SLR. 1. F-Test: Tests for a significant relationship between the dependent variable and the set of all independent variables. We refer to this as the test for overall significance. 2. If the F-Test shows overall significance, than we use the t- test to check the significance for each of the independent variables. We conduct the t-test on each of the independent variables. We refer to the t-test as the test for individual significance.
  • 9. F-Test In multiple regression, we test that none of the parameters are equal to zero: H0: β0 = β1 = ... = βp = 0 Ha: One or more of the parameters are equal to zero. Remember that F = MSR/MSE. And in multiple regression: MSR = SSR/p MSE = SSE/(n - p - 1) And we reject H0 if our p-value < α
  • 10. T-Test Remember we test for each parameter. For any βi: H0: βi = 0 Ha: βi ≠ 0 t = βi/sbi And we reject H0 if our p-value < α
  • 11. Multicollinearity This is essentially the correlation among independent variables. We care about this because we want our independent variables to measure significantly different things when predicting our dependent variable. While in practice there is always some multicollinearity, we need to try and eliminate as much as we can. A simple test of multicollinearity is with the sample correlation (rx1x2) for any two independent variables. If the sample correlation exceeds .7 for any two independent variables we must take measures to reduce multicollinearity, for example, removing one of the two highly correlated variables from the model.
  • 12. The End Thats it for Ch. 15. Hope you have recovered from Mardi Gras next time I see you!