[R] Capture of iterative integration output

Terry W. Schulz terrywschulz at comcast.net
Sun Aug 27 00:42:17 CEST 2006


I am a novice R user and am having difficulty retrieving  the values  
from 21 iterations of the R function integrate.
The only way I have found is to do a write.table and then a read.table 
as shown in the code below.  I would rather capture the 21 values inside 
the braces ( sapply might work, but I can't set it up without getting an 
error in function) so I could compute the negative log-likelihood inside 
the braces.  My goal is to then use optim to find the best fitting mu 
and sigma.  In the code that follows mu and sigma are given, but 
normally they would not be known.

Any help on y$value capture would be appreciated.  I would be ecstatic 
for help on implementation of the optim function using finite 
differences for the following code. 

By the way in case anyone is interested function(x) is the 
Poisson-lognormal pdf.

q <- read.table(file="c:/Bayes/FigA3-1.prn",header=TRUE)
V <- c(volume)
c <- c(count)
n <- length(count)
mu <- -8.202900475
sigma <- 1.609437912
for(i in c(1:n))
integrand <- function(x) 
y <- integrate(integrand, lower = 0, upper = Inf)
z <-read.table(file="c:/Bayes/PLNA3-1.out")
z <- c(z)
LL <- log(c(z$V1))
#negative Log-Likelihood

Terry W. Schulz
Applied Statistician
1218 Pomegranate Lane
Golden, CO 80401

terrywschulz at comcast.net
(303) 526-1461

More information about the R-help mailing list