Hey fellas:
In the context of the gambler's ruin problem, the following R code obtains the
mean duration of the game, in turns:
# total.capital is a constant, an arbitrary positive integer
# initial.capital is a constant, an arbitrary positive integer between, and not
including
# 0 and total.capital
# p is the probability of winning 1$ on each turn
# 1-p is the probability of loosing 1$
# N is a large integer representing the number of times to simulate
# dur is a vector containing the simulated game durations
T <- total.capital
dur <- NULL
for (n in 1:N) {
x <- initial.capital
d <- 0
while ((x!=0)&(x!=T)) {
x <- x+sample(c(-1,1),1,replace=TRUE,c(1-p,p))
d <- d+1
}
dur <- c(dur,d)
}
mean(dur) #returns the mean duration of the game
The problem with this code is that, using the traditional control structures
(while, for, etc.) it is rather slow. Does anyone know of a way i could
vectorize the "while" and the "for" to produce a faster code?
And while I'm at it, does anyone know of a discrete-event simulation package in
R such as the "SimPy" for Python?
Thanks in advance
[[alternative HTML version deleted]]
______________________________________________
[email protected] 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.
