Hi all,
I have bootstrapped my weighted regression equation using the
following formula and the package boot to obtain bootstrap confidence
intervls for my parameters:
>drought <- read.table("D:/drought080525.txt", header=T)
>regres <- function(x, indices) {
>x <- x[indices,]
>coef(lm(x$AGB ~ x$days, weights=x$weights))
>}
>bst <- boot(drought, regres, R=10000, stype="i")
ORDINARY NONPARAMETRIC BOOTSTRAP
Call:
boot(data = drought, statistic = regres, R = 10000, stype = "i")
Bootstrap Statistics :
original bias std. error
t1* 0.4186985 0.010107259 0.6995259
t2* -0.4334417 0.001973715 0.1053180
>AGB_bstci1 <- boot.ci(bst, conf=0.95, type="bca",index=1)
>AGB_bstci2 <- boot.ci(bst, conf=0.95, type="bca",index=2)
BOOTSTRAP CONFIDENCE INTERVAL CALCULATIONS
Based on 10000 bootstrap replicates
CALL :
boot.ci(boot.out = bst, conf = 0.95, type = "bca", index = 1)
Intervals :
Level BCa
95% (-1.2767, 1.5678 )
Calculations and Intervals on Original Scale
BOOTSTRAP CONFIDENCE INTERVAL CALCULATIONS
Based on 10000 bootstrap replicates
CALL :
boot.ci(boot.out = bst, conf = 0.95, type = "bca", index = 2)
Intervals :
Level BCa
95% (-0.5977, -0.1661 )
Calculations and Intervals on Original Scale
I would like to plot my regression line and the bootstrapped
confidence bands, but am a bit at a loss on how to convert the
bootstrapped confidence intervals for the parameters into a confidence
band for the regression line. Does anyone any ideas?
All help is greatly apprecited!
Many thanks,
Geertje
______________________________________________
R-help@r-project.org mailing list
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.
