# distributions: Wrappers to random number generators for use with coenocliner In coenocliner: Coenocline Simulation

## Description

These functions are simple wrappers around existing random number generators in R to provide stochastic count data for simulated species.

## Usage

 ``` 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17``` ```NegBin(n, mu, alpha) Poisson(n, mu) Bernoulli(n, mu) Binomial(n, mu, size) BetaBinomial(n, mu, size, theta) ZIP(n, mu, zprobs) ZINB(n, mu, alpha, zprobs) ZIB(n, mu, size, zprobs) ZIBB(n, mu, size, theta, zprobs) ```

## Arguments

 `n` the number of random draws, equal to number of species times the number of gradient locations. `mu` the mean or expectation of the distribution. For `Bernoulli`, `Binomial`, and `BetaBinomial()` this is the probability of occurrence as given by the response function. `alpha` numeric; dispersion parameter for the negative binomial distribution. May be a vector of length `length(mu)`. The NB2 parametrization of the negative binomial is used here, in which α is positively related to the amount of extra dispersion in the simulated data. As such, where α = 0, we would have a Poisson distribution. `alpha` can be supplied a value of `0`, in which case `NegBin` and `ZINB` return random draws from the Poisson or zero-inflated Poisson distributions, respectively. Negative values of `alpha` are not allowed and will generate an error. `size` numeric; binomial denominator, the total number of individuals counted for example `theta` numeric; a positive inverse overdispersion parameter for the Beta-Binomial distribution. Low values give high overdispersion. The variance is `size*mu*(1-mu)*(1+(size-1)/(theta+1))` (Bolker, 2008) `zprobs` numeric; zero-inflation parameter giving the proportion of extraneous zeros. Must be in range 0 to 1.

## Value

a vector of random draws from the stated distribution.

Gavin L. Simpson

## References

Bolker, B.M. (2008) Ecological Models and Data in R. Princeton University Press.

coenocliner documentation built on May 29, 2017, 3:57 p.m.