Study Guides (273,580)
AUS (8,030)
ANU (260)
STAT (12)
STAT1008 (12)
Bronwyn (12)
Final

STAT1008 Study Guide - Stepwise Regression, Confounding, Dependent And Independent Variables

by OneClass2447024 , Summer 2017
3 Pages
77 Views

Department
Statistics
Course Code
STAT1008
Professor
Bronwyn
Study Guide
Final

This preview shows page 1. Sign up to view the full 3 pages of the document.
Multiple Regression
Response variable: Y (quantitative)
k predictors: X1, X2, … Xk (quantitative or 0/1 categories)
Model: Y = β0 1X1 2X2 +···kXk
ε ~ N0,σε) and independent.
What to do?
Estimate the coefficients: b0, b1,…, k (estimate the same way as a simple model).
Test the individual predictors: t-tests (use ŷ?.
Assess the overall fit: R2, ANOVA.
T-tests for Individual Predictors:
Y = β0 1X1 +β2X2 +···+βkXk
Testing whether X2 be included in the model or not, eg: should exam 2 be in the model
given the presence of exam 1.
i.e. Testing the effectiveness of any predictor, say Xi, in a multiple regression model.
H0: βi = 0 vs Ha: βi 0
Test statistic: t = bi/Sebi
We find a p-value using a t-distribution with n - k - 1 df, where k is the number of
predictors.
If we reject the null we see that the predictor is an effective contributor to this model.
Individual t-tests assess the importance of a predictor after the other predictors are
already in the model, eg: if I add another predictor into the model, exam 1 may
eoe isigifiat.
Assessing Overall Fit: R2
R2 = % of ariailit i Y hih is eplaied  the odel.
SSTotal =  (as for one predictor)
SSE =   (as for one predictor)
SSModel = SSTotal SSE
R2 = SSModel/SSTotal
Adjusted R2 usually lower as it contains an additional value.
R2 is interpreted as the percent of variability in the response values in the sample that
is explained by the fitted regression model.
Eg: If R2 =0.525, this means that 52.5% of the variability in Final exam scores is
explained by the model based on exam 1 and exam 2.
Assessing Overall Fit: ANOVA
To test for the overall effectiveness of a regression model: Y = β0 + β1X1 2X2 +···+βkXk
H0: βi = β2 =…= βk = 0 (the model is ineffective)
Ha: At least oe βi 0 (at least one predictor in the model is effective)
Mean square = SS/df
10.2 Checking Regression Conditions
Conditions for a Regression Model:
find more resources at oneclass.com
find more resources at oneclass.com

Loved by over 2.2 million students

Over 90% improved by at least one letter grade.

Leah — University of Toronto

OneClass has been such a huge help in my studies at UofT especially since I am a transfer student. OneClass is the study buddy I never had before and definitely gives me the extra push to get from a B to an A!

Leah — University of Toronto
Saarim — University of Michigan

Balancing social life With academics can be difficult, that is why I'm so glad that OneClass is out there where I can find the top notes for all of my classes. Now I can be the all-star student I want to be.

Saarim — University of Michigan
Jenna — University of Wisconsin

As a college student living on a college budget, I love how easy it is to earn gift cards just by submitting my notes.

Jenna — University of Wisconsin
Anne — University of California

OneClass has allowed me to catch up with my most difficult course! #lifesaver

Anne — University of California
Description
Multiple Regression Response variable: Y (quantitative) k predictors: X1, X2, Xk(quantitative or 0/1 categories) Model: Y = 0 X 1 1X +2 2+ X +k k ~ N(0,) and independent. What to do? Estimate the coefficients: b ,0b 1, b kestimate the same way as a simple model). Test the individual predictors: t-tests (use ?). 2 Assess the overall fit: R , ANOVA. T-tests for Individual Predictors: Y = 0+ 1 1 X 2 2+ X +k k Testing whether X b2 included in the model or not, eg: should exam 2 be in the model given the presence of exam 1. i.e. Testing the effectiveness of any predictor, say X,iin a multiple regression model. H0: i= 0 vs Ha: i 0 Test statistic: t =ib/Sbi We find a p-value using a t-distribution with n - k - 1 df, where k is the number of predictors. If we reject the null we see that the predictor is an effective contributor to this model. Individual t-tests assess the importance of a predictor after the other predictors are already in the model, eg: if I add another predictor into the model, exam 1 may become insignificant. Assessing Overall Fit: R 2 2 R = % of variability in Y which is explained by the model. SSTotal = ( ) (as for one predictor) 2 SSE = ( ) (as for one predictor) SSModel = SSTotal SSE 2 R = SSModel/SSTotal Adjusted R usually lower as it contains an additional value. R is interpreted as the percent of variability in the response values in the sample that is explained by the fitted regression model. Eg: If R =0.525, this means that 52.5% of the variability in Final exam scores is explained by the model based on exam 1 and exam 2. Assessing Overall Fit: ANOVA To test for the overall effectiveness of a regression model: Y = + 0 X +1 1 +2 2 X k k + H0: i= 2== =k0 (the model is ineffective) Ha: At least one i 0 (at least one predictor in the model is effective) Mean square = SS/df 10.2 Checking Regression Conditions Conditions for a Regression Model:
More Less
Unlock Document
Start your 2 week free trial

Only page 1 are available for preview. Some parts have been intentionally blurred.

Unlock Document
Start your 2 week free trial
You're Reading a Preview

Unlock to view full version

Unlock Document
Start your 2 week free trial

Share with your friends

Get 2 friends to sign-up for a free trial as well, and get an additional free week

Next

Log In


OR

Don't have an account?

Join OneClass

Access over 10 million pages of study
documents for 1.3 million courses.

Sign up

Join to view


OR

By registering, I agree to the Terms and Privacy Policies
Already have an account?
Just a few more details

So we can recommend you notes for your school.

Reset Password

Please enter below the email address you registered with and we will send you a link to reset your password.

Add your courses

Get notes from the top students in your class.


Submit