Displaying 1 result from an estimated 1 matches for "fitw".
Did you mean:
fit
2010 Jul 27
1
problem with zero-weighted observations in predict.lm?
...8 [1] 3
$residual.scale $residual.scale
[1] 0.1031362 [1] 0.1684207
### start of script ###
data <- data.frame(x=1:10, y=log(1:10))
fitw <- lm(data=data, y~x, weights=as.numeric(x<=5))
fits <- lm(data=data, y~x, subset=x<=5)
fitw$df.residual == 3 && fits$df.residual == 3 # TRUE
identical(coef(fitw), coef(fits)) # TRUE
sumw <- summary(fitw)
sums <- summary(fits)
identical(sumw$df, sums$df) # TRUE
predw &l...