Linear Model and Binomial Mixed Data Type Distribution

Density, and random generation for the (four parameter bivariate) Linear Model–Bernoulli copula distribution.

Usage

dN1binom(x1, x2, mean = 0, sd = 1, prob, apar = 0,
         copula = "gaussian", log = FALSE)
rN1binom(n, mean = 0, sd = 1, prob,
         apar = 0, copula = "gaussian")

Arguments

x1, x2

vector of quantiles. The valid values of x2 are \(0\) and \(1\).

n

number of observations. Same as rnorm.

copula

See N1binomial.

mean, sd, prob, apar

See N1binomial.

log

Logical. If TRUE then the logarithm is returned.

Value

dN1binom gives the probability density/mass function, rN1binom generates random deviate and returns a two-column matrix.

Author

T. W. Yee

Details

See N1binomial, the VGAM family functions for estimating the parameter by maximum likelihood estimation, for details.

See also

N1binomial, rnorm, rbinom.

Examples

if (FALSE) { # \dontrun{ 
nn <- 1000;  apar <- rhobitlink(1.5, inverse = TRUE)
prob <- logitlink(0.5, inverse = TRUE)
mymu <- 1; sdev <- exp(1)
mat <- rN1binom(nn, mymu, sdev, prob, apar)
bndata <- data.frame(y1 = mat[, 1], y2 = mat[, 2])
with(bndata, plot(jitter(y1), jitter(y2), col = "blue"))
} # }