I have a few questions about using add1(). First of all, according to the R
document of add1(), the trace = TRUE function prints out progress reports
but my attempts to do this have failed several times. So how do we print
out the progress reports of adding terms to the model so as to test for
improved fit (or not)? Also, why the LRT (likelihood-ratio test) value is
not in the output? Any help is greatly appreciated! Example as follows:
require(graphics); require(utils)
lm1 <- lm(Fertility ~ ., data = swiss)
add1(lm1, scope = .~. + .^2, test="Chisq", trace = TRUE)
#It seems that no progress report has been printed. Also `LRT` is missing
Single term additions
Model:
Fertility ~ Agriculture + Examination + Education + Catholic +
Infant.Mortality
Df Sum of Sq RSS AIC Pr(>Chi)
<none> 2105 191
Agriculture:Examination 1 10.7 2094 192 0.6251
Agriculture:Education 1 1.8 2103 193 0.8399
Agriculture:Catholic 1 75.0 2030 191 0.1915
Agriculture:Infant.Mortality 1 4.4 2101 193 0.7528
Examination:Education 1 48.7 2056 192 0.2943
Examination:Catholic 1 40.8 2064 192 0.3378
Examination:Infant.Mortality 1 65.9 2039 191 0.2216
Education:Catholic 1 278.2 1827 186 0.0099 **
Education:Infant.Mortality 1 93.0 2012 191 0.1451
Catholic:Infant.Mortality 1 2.4 2103 193 0.8184
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Best,
Chao
[[alternative HTML version deleted]]
______________________________________________
R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
