InfluentialPoints.com
Biology, images, analysis, design...
Use/Abuse Principles How To Related
"It has long been an axiom of mine that the little things are infinitely the most important" (Sherlock Holmes)

Search this site

 

 

Just a note

Using R
hosp=read.table('climent2.txt',header=T) > mod1=lm(manh~load+xray+beds+popn+stay,data=hosp) > library(car) > vif(mod1) load xray beds popn stay 462.960478 29.917368 434.014426 15.273862 5.650849 > # REMOVE LOAD AS HAS THE LARGEST VIF > 5 > mod2=update(mod1,~.-load) > library(car) > vif(mod2) xray beds popn stay 29.207164 34.822404 10.215777 5.507806 > # REMOVE BEDS AS HAS THE LARGEST VIF >5 > mod3=update(mod2,~.-beds) > vif(mod3) xray popn stay 9.969402 10.148112 1.236187 > # REMOVE POPN AS HAS THE LARGEST VIF >5 > mod4=update(mod3,~.-popn) > vif(mod4) xray stay 1.213956 1.213956 > #ALL VIF < 5 SO EXAMINE MODEL > summary(mod4) Call: lm(formula = manh ~ xray + stay, data = hosp) Residuals: Min 1Q Median 3Q Max -1707.1 -462.5 39.2 552.7 1264.5 Coefficients: Estimate Std. Error t value Pr(>|t|) (Intercept) -2.746e+03 8.624e+02 -3.184 0.00718 ** xray 2.105e-01 1.166e-02 18.052 1.38e-10 *** stay 6.058e+02 1.542e+02 3.928 0.00173 ** --- Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 Residual standard error: 871 on 13 degrees of freedom Multiple R-squared: 0.974, Adjusted R-squared: 0.97 F-statistic: 243.3 on 2 and 13 DF, p-value: 5.008e-11