multivariate linear regression

by Trilli@n
Tags: linear, multivariate, regression
Trilli@n is offline
May22-10, 03:02 AM
P: 2
I have a model y= beta0 + beta1 x1 + beta2 x2 + eps, eps~N(0,1).

How to test hypothesis beta1=0 ? Is the same test for beta2=0?
Phys.Org News Partner Science news on
SensaBubble: It's a bubble, but not as we know it (w/ video)
The hemihelix: Scientists discover a new shape using rubber bands (w/ video)
Microbes provide insights into evolution of human language
Damned charming :)
Damned charming :) is offline
May22-10, 06:35 AM
P: 66
you can use a t test just like you do in simple linear regression,
except the degrees of freedom is n-3 not n-2,
You can test each variable separately.

you have to be careful about any conlcusions you make for example if x1 and x2 are highly correlated.
Trilli@n is offline
May31-10, 05:36 PM
P: 2
Can someone help me to do this in SAS?
Is this correct?

PROC REG DATA = dataset;
MODEL y = x1 x2;
test x1=0 ;
test x2=0;

How can I save p value from each test to a new data?

Register to reply

Related Discussions
Regression SS in multiple linear regression Set Theory, Logic, Probability, Statistics 2
Multivariate regression for project cost estimates Set Theory, Logic, Probability, Statistics 6
Linear regression in R Programming & Computer Science 3
[SOLVED] Multivariate Linear Regression With Coefficient Constraint Set Theory, Logic, Probability, Statistics 5