There are several mistakes:
Why are you using only p[1], p[2]?
Afterwards you try (for i in 1:10) p[i] so you should go to p[10]?
Then you should make a pnew wich holds 10 values, so pnew[i]
This should get you back on track
Bart
Kathy Maher wrote:
>
> Hi,
>
> I am trying to use the Fisher scoring method with a geometric
> distribution,
> with p = .07, 100 observations from the geom distrib, and 10 iterations.
> I cannot quite get the code to work.
> Can anyone see the mistake?
>
>
>
>
> n <- 100
>
> p <- 0.07
>
> x <- rgeom(n, p)
>
> s <- sum(x)
>
> f <- function(x, p) p*(1-p)^x
>
> L <- function(p) p^n*(1-p)^s
>
> logL <- function(p) n*log(p)+s*(log(1-p))
>
> logLprime <- function(p) (n/p)-(s/(1-p))
>
>
>
> I <- n/p^2*(1-p)
>
> iter <- 10
>
> p[1] <- .06
>
> p[2] <- .11
>
>
>
>
>
> for (i in 1:10)
>
> {
>
> pnew <- p[i]+logLprime(p[i])/I*(p[i])
>
> }
>
> [[alternative HTML version deleted]]
>
> ______________________________________________
> 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.
>
>
--
View this message in context:
http://www.nabble.com/%28no-subject%29-tp16084593p16093575.html
Sent from the R help mailing list archive at Nabble.com.
______________________________________________
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.
