We are trying to randomly sample from a right skewed distribution with a mean of 23, low 95% CI 18 and high of 36. Any examples will be greatly appreciated. Whole numbers only.

This distribution is also binomial but when I use rbinom, I just get weird numbers (and I don't know the probability of any given number between 18 and 36 necessarily).

First, thank you so much for your response!

We used program MARK after capturing 14 individuals in a particular trapping session to estimate the actual number of individuals at 23 with 95% confidence interval of 18-36. We need to run this on multiple trapping sessions with different numbers of individuals captured and different CIs for each one.

The probability of that number being 23 is unknown and when I used binom functions, I just guessed to see if it would give us anything of use. We need the probability of every number between 18 and 36 so that we can use this in an energetics model sampling from this distribution each time we run 10,000 MCMC replicates.

I have tried:

q <- seq(18:36)

pbinom(q, 23, 0.6)

and got probabilities that make no sense because 36 is more likely than 23.

When I try

rbinom(100, 23, 0.6)

I get a bunch of numbers lower than 23 and certainly lower than 18 so it's not giving me the distribution of numbers I am looking for.

I have tried using the package: sn but my numbers are out of the feasible range for the cp2dp command.

Does that help clarify things? Again, thank you so much for your help!

Just so future people that might be looking for something similar can find it:

A good approximation of the distribution we are looking for is found by using rlnorm. Someone way smarter than me figured it out.

This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.

If you have a query related to it or one of the replies, start a new topic and refer back with a link.