Dear R users,
I am trying to simulate surveys and the survey result will be used to
determine the population to be "accepted" or "rejected".
With the results,
I would like to calculate cumulative means and plot them to see if a
converged value is as expected. Below is R-code I generated. I need a
help to repeat this simulation code as many times (e.g., 100) and keep the
results as list format if possible. Could you give me any insight?
Thanks a lot in advance,
Steve
sim.f <- function(p.s, N, sample.size, n.sim) {
pop = sampled.pop = decision = decisionB = cum.mn = as.list(NULL)
for(i in 1:n.sim) {
p <- c(rep(1, p.s*N), pop2 <- rep(0, N*(1-p.s))) # Generate sample
space
pop[[i]] <- sample(p) # Randomization sample space
sampled.pop[[i]] <- sample(pop[[i]], sample.size)# Random sampling
decision[i] <- ifelse(sum(sampled.pop[[i]])>=1,
'Reject','Pass') #
Decision for each group of n.sim
decisionB <- ifelse(decision == 'Reject', 1, 0) # Convert to
binary
cum.mn <- cumsum(decisionB) / seq_along(decisionB) # Cummulative mean of
n.sim group decisions
}
result = list(population=pop,
pop_sub = sampled.pop,
decision = decision,
decisionB = decisionB,
cum.mn = cum.mn)
}
sim.out <- sim.f(p.s=.05, N=1000, sample.size=69, n.sim=500)
# I want to repeat this simulation function for example 100 times or and
also #keep the data so that I can explore later. If it is not possible to
keep all #outputs, at least I would like to have cum.mn outputs.
summary(sim.out)
sim.out$population
sim.out$pop_sub
sim.out$decision
sim.out$decisionB
y1 <- sim.out$cum.mn
#plot(y1, type='l')
lines(y2, type='l')
...
lines(y100, type='l')
abline(h=.95, col='red')
[[alternative HTML version deleted]]
Here's one way to save your results, using a list of lists and a for() loop.
nsim <- 100
outputs <- vector("list", nsim)
for(i in 1:nsim) {
outputs[[i]] <- sim.f(p.s=.05, N=1000, sample.size=69, n.sim=500)
}
Jean
On Fri, Sep 18, 2015 at 2:27 PM, SH <emptican at gmail.com> wrote:
> Dear R users,
>
> I am trying to simulate surveys and the survey result will be used to
> determine the population to be "accepted" or
"rejected". With the results,
> I would like to calculate cumulative means and plot them to see if a
> converged value is as expected. Below is R-code I generated. I need a
> help to repeat this simulation code as many times (e.g., 100) and keep the
> results as list format if possible. Could you give me any insight?
>
> Thanks a lot in advance,
>
> Steve
>
> sim.f <- function(p.s, N, sample.size, n.sim) {
> pop = sampled.pop = decision = decisionB = cum.mn = as.list(NULL)
> for(i in 1:n.sim) {
> p <- c(rep(1, p.s*N), pop2 <- rep(0, N*(1-p.s))) # Generate sample
space
> pop[[i]] <- sample(p) # Randomization sample space
> sampled.pop[[i]] <- sample(pop[[i]], sample.size)# Random sampling
> decision[i] <- ifelse(sum(sampled.pop[[i]])>=1,
'Reject','Pass') #
> Decision for each group of n.sim
> decisionB <- ifelse(decision == 'Reject', 1, 0) # Convert to
binary
> cum.mn <- cumsum(decisionB) / seq_along(decisionB) # Cummulative mean
> of
> n.sim group decisions
> }
> result = list(population=pop,
> pop_sub = sampled.pop,
> decision = decision,
> decisionB = decisionB,
> cum.mn = cum.mn)
> }
> sim.out <- sim.f(p.s=.05, N=1000, sample.size=69, n.sim=500)
> # I want to repeat this simulation function for example 100 times or and
> also #keep the data so that I can explore later. If it is not possible to
> keep all #outputs, at least I would like to have cum.mn outputs.
>
> summary(sim.out)
> sim.out$population
> sim.out$pop_sub
> sim.out$decision
> sim.out$decisionB
> y1 <- sim.out$cum.mn
> #plot(y1, type='l')
> lines(y2, type='l')
> ...
> lines(y100, type='l')
> abline(h=.95, col='red')
>
> [[alternative HTML version deleted]]
>
> ______________________________________________
> R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see
> 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.
>
[[alternative HTML version deleted]]
Hi Jean, Thank you so much! Steve On Sat, Sep 19, 2015 at 1:02 PM, Adams, Jean <jvadams at usgs.gov> wrote:> Here's one way to save your results, using a list of lists and a for() > loop. > > nsim <- 100 > outputs <- vector("list", nsim) > for(i in 1:nsim) { > outputs[[i]] <- sim.f(p.s=.05, N=1000, sample.size=69, n.sim=500) > } > > Jean > > On Fri, Sep 18, 2015 at 2:27 PM, SH <emptican at gmail.com> wrote: > >> Dear R users, >> >> I am trying to simulate surveys and the survey result will be used to >> determine the population to be "accepted" or "rejected". With the >> results, >> I would like to calculate cumulative means and plot them to see if a >> converged value is as expected. Below is R-code I generated. I need a >> help to repeat this simulation code as many times (e.g., 100) and keep the >> results as list format if possible. Could you give me any insight? >> >> Thanks a lot in advance, >> >> Steve >> >> sim.f <- function(p.s, N, sample.size, n.sim) { >> pop = sampled.pop = decision = decisionB = cum.mn = as.list(NULL) >> for(i in 1:n.sim) { >> p <- c(rep(1, p.s*N), pop2 <- rep(0, N*(1-p.s))) # Generate sample >> space >> pop[[i]] <- sample(p) # Randomization sample space >> sampled.pop[[i]] <- sample(pop[[i]], sample.size)# Random sampling >> decision[i] <- ifelse(sum(sampled.pop[[i]])>=1, 'Reject','Pass') # >> Decision for each group of n.sim >> decisionB <- ifelse(decision == 'Reject', 1, 0) # Convert to binary >> cum.mn <- cumsum(decisionB) / seq_along(decisionB) # Cummulative mean >> of >> n.sim group decisions >> } >> result = list(population=pop, >> pop_sub = sampled.pop, >> decision = decision, >> decisionB = decisionB, >> cum.mn = cum.mn) >> } >> sim.out <- sim.f(p.s=.05, N=1000, sample.size=69, n.sim=500) >> # I want to repeat this simulation function for example 100 times or and >> also #keep the data so that I can explore later. If it is not possible to >> keep all #outputs, at least I would like to have cum.mn outputs. >> >> summary(sim.out) >> sim.out$population >> sim.out$pop_sub >> sim.out$decision >> sim.out$decisionB >> y1 <- sim.out$cum.mn >> #plot(y1, type='l') >> lines(y2, type='l') >> ... >> lines(y100, type='l') >> abline(h=.95, col='red') >> >> [[alternative HTML version deleted]] >> >> ______________________________________________ >> R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see >> 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. >> > >[[alternative HTML version deleted]]