This preview has intentionally blurred sections. Sign up to view the full version.View Full Document
Unformatted text preview: # R code: gas consumption data STAT 331 Tutorial, Feb.5, 2010 # multiple regression and hypothesis tests # example data: example 1.2.7 on page11 in Introduction to regression modeling gas=read.table(’E:/gas.txt’,head=T) attach(gas) # read the data set cor(gas) # correlation matrix for ALL variables in data set # gives correlation r, measure of linear relationship between each pair of variables pairs(gas) # look for association between: # 1. response variable and any of predictor variables # 2. any two predictor variables model=lm(MPG ∼ WT+DIS+NC+HP+ACC+ET,data=gas) summary(model) # gives: estimation of regression coefficients, standard errors, t-statistics # here, t k is for testing # H : β k = 0 # H 1 : β k negationslash = 0 # also gives: r 2 , F-statistic # here, F is for testing regression relationship: # H : β 1 = β 2 = ··· = β k = 0 # H 1 : not all β k negationslash = 0 # F α ( p, n − p − 1) = F . 05 (6 , 31) = 2 . 42, when F > 2 . 42, refuse H # t α/ 2 ( n −...
View Full Document
This note was uploaded on 01/20/2011 for the course MATH 332 taught by Professor Jt during the Spring '10 term at Waterloo.
- Spring '10