nullTest: Simulation based test for null-hypothesis, H0:theta=0
Description
Simulates data sets under the null-hypothesis,
$H_0:\theta=0$. This corresponds to an ordinary multinomial
model without any overdispersion. Based on the returned data frame
simulated $p$-values may be computed.
Usage
nullTest(data, m=1000, prec=6)
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.
m
Number of simulated data tables.
prec
The tolerance of the iterations. Corresponds to
epsilon=1e-prec in dirmult.
Value
Returns a data frame with theta estimates and log-likelihood values.