gridProf: Profile log-likelihood of Dirichlet-multinomial model

In dirmult: Estimation in Dirichlet-Multinomial Distribution

Profile log-likelihood of Dirichlet-multinomial model

Description

Computes the profile log-likelihood of l(π,theta;x) for a given sequence of theta by calling estProfLogLik.

Usage

gridProf(data, theta, from, to, len)

Arguments

data	A matrix or table with counts. Rows represent subpopulations and columns the different categories of the data. Zero rows or columns are automaticly removed.
theta	A theta-value used as offset for the interval: [theta+from; theta+to].
from	Left endpoint in the interval: [theta+from; theta+to].
to	Right endpoint in the interval: [theta+from; theta+to].
len	Number of points in the [from; to] interval. Similar to the len argument in seq.

Value

Gives a data frame with theta values and associated profile log-likelihood values.

See Also

estProfLogLik

Examples

data(us)
fit <- dirmult(us[[1]],epsilon=10^(-12),trace=FALSE)
## Not run: grid <- gridProf(us[[1]],fit$theta,from=-0.001,to=0.001,len=10)
plot(loglik ~ theta, data=grid, type="l")
## End(Not run)

dirmult documentation built on March 21, 2022, 5:05 p.m.