Fox Module 12: Statistical inference for multiple regression HW


Fox Module 12: Statistical inference for multiple regression HW

Author
Message
NEAS
Supreme Being
Supreme Being (6K reputation)Supreme Being (6K reputation)Supreme Being (6K reputation)Supreme Being (6K reputation)Supreme Being (6K reputation)Supreme Being (6K reputation)Supreme Being (6K reputation)Supreme Being (6K reputation)Supreme Being (6K reputation)

Group: Administrators
Posts: 4.5K, Visits: 1.6K

Module 12: Statistical inference for multiple regression

 

(The attached PDF file has better formatting.)

 

Homework assignment: F test and analysis of variance

 

This homework assignment continues the scenario in Module 9.

 

We regress the Y values on the X1 and X2 values in the table below.

 

X1

X2

Y

X1

X2

Y

X1

X2

Y

X1

X2

Y

1

1

-0.395

1

2

-1.705

1

3

-2.942

1

4

-3.634

2

1

1.942

2

2

0.964

2

3

-2.463

2

4

-1.349

3

1

1.717

3

2

0.206

3

3

0.397

3

4

-0.982

4

1

2.258

4

2

2.908

4

3

-0.092

4

4

-0.235

 


A.     What is the null hypothesis for the omnibus F test?

B.     What are the total sum of squares (TSS), regression sum of squares (ResSS), and residual sum of squares (RSS)?

C.    What are the degrees of freedom for the residual sum of squares and regression sum of squares?

D.    What is the value of the F statistic?

E.     What is the p value for this F statistic?

 


 

A.     Show the formulas and the computations for Parts A through D.

B.     Use Excel or other statistical software to find the p value in Part E.


 

 


Attachments
Fox Module 12 F test ANOVA HW.pdf (1.9K views, 34.00 KB)
Reply
Michelle2010
Junior Member
Junior Member (19 reputation)Junior Member (19 reputation)Junior Member (19 reputation)Junior Member (19 reputation)Junior Member (19 reputation)Junior Member (19 reputation)Junior Member (19 reputation)Junior Member (19 reputation)Junior Member (19 reputation)

Group: Forum Members
Posts: 18, Visits: 1

I need some help on part E.  I'm not sure how to get the p-value for this hypothesis test, and unless I'm confused, I don't believe there is anyway to get it out of the regression option in Excel's analysis toolpak.

The only other way I see to use in Excel for calculating p-values needs the t-value as an input.  It is TDIST(t-statistic, degrees of freedom, number of tails).

Using the formula at the bottom of page 109, F0= {(RegSS1 - RegSS0)/q}/{RSS1/(n-k-1)}, I calculated F0=33.29.

I see in the solution to practice problem 1.2 from "Module 12: Statistical Inference for Multiple Linear Regression" that "in the two-variable regression model, the t statistic is the square root of the F ratio."  I therefore concluded that the t-statistic = +- the square root of 33.29 = +-5.7697.  I calculated the degrees of freedom as n-k-1 = 16-2-1=13.  I then computed the p-value by typing TDIST = (5.7697, 13, 2) in excel : .0001. 

Did I approach the problem correctly?  Also, why does excel give me an error when I attempt to enter the negative t-statistic (TDIST = (-5.7697, 13, 2))?

[NEAS: What you did is correct for one explanatory variable. It is simpler to use the regression add-in, which give p-values for coefficients and for F statistics. You need the regression add-in to easily do multiple regression for the student project. Click on the Tools menu, then data analysis, then regression.]

 

 


GO
Merge Selected
Merge into selected topic...



Merge into merge target...



Merge into a specific topic ID...






Reading This Topic


Login
Existing Account
Email Address:


Password:


Social Logins

  • Login with twitter
  • Login with twitter
Select a Forum....













































































































































































































































Neas-Seminars

Search