Internal function

These functions estimate the parameters of the (zero-inflated) negative binomial distribution by applying the maximum likelihood method to the labelled observations in class 0.

estim.nbinom(y, z, gamma)

estim.zinb(y, z, gamma)

Arguments

y	observations: numeric vector of length n
z	class labels: integer vector of length n, with entries 0, 1 and NA
gamma	offset: numeric vector of length n, or NULL

Value

These functions return a list of numerics.

See also

These are internal functions. The user functions are mixtura and scrutor.

Examples

# data simulation
n <- 100
y <- stats::rnbinom(n=n,mu=5,size=1/0.05)
y[sample(1:n,size=0.2*n)] <- 0
z <- rep(0,times=n)
gamma <- rep(1,times=n)

# parameter estimation
estim.nbinom(y,z,gamma)
#> $mu
#> [1] 4.11
#> 
#> $phi
#> [1] 0.5959951
#> 
estim.zinb(y,z,gamma)
#> $mu
#> [1] 5.014082
#> 
#> $phi
#> [1] 0.1517328
#> 
#> $pi
#> [1] 0.1803093
#>