GenerateProbNumbers: Sample positions from a given probability density...
In DavidGarciaCallejas/DGC: David Garcia-Callejas' functions

Description Usage Arguments Value Author(s) Examples

this function randomly selects "times" positions in a vector according to a given distribution, either "power-law" or provided in the column "my.data$prob" It accepts a power-law distribution: y = c*x^k this parameterization is consistent with Hatton et al. 2015, who found a general scaling of k ~ 0.75 por prey and predator biomass across ecosystems globally hence the default k = 0.75. it also accepts a lognormal distribution with logmean and logsd and a gambin distribution with parameters gambin.alpha and gambin.maxoctave

1
2
3

GenerateProbNumbers(times, choice.length = 100, dist = "power-law", c = 1,
  k = 0.75, logmean = 0, logsd = 0.5, gambin.alpha = 2,
  gambin.maxoctave = 8, cum.sum = 0, my.data = NULL)

`times`	number of positions to be selected
`choice.length`	length of the vector to select positions from
`dist`	either "uniform", "power-law", "lognormal" or "gambin"
`c`	for the power law parametrization
`k`	exponent of the power law
`logmean`	mean of the lognormal distribution
`logsd`	standard deviation of the lognormal distribution
`gambin.alpha`	alpha value of the gambin distribution
`gambin.maxoctave`	value of the maximum octave of the gambin distribution -TODO-
`cum.sum`	cumulative sum of the returning values
`my.data`	dataframe specifying the distribution to sample from, instead of dist. Of the form my.data$prob

numeric vector of length times

David Garc<c3><ad>a-Callejas, david.garcia.callejas@gmail.com

t1 <- GenerateProbNumbers(100,c = 1, k = 0.75,cum.sum = 100)
hist(t1)
table(t1)
t2 <- GenerateProbNumbers(times = 2000, k = 0.25, cum.sum = 10000)
t3 <- GenerateProbNumbers(times = 80, choice.length = 100, dist = "gambin", gambin.alpha = 2, gambin.maxoctave = 8, cum.sum = 10000)
hist(t2)
hist(t3)