Set Control Parameters for emCgm

DESCRIPTION:

Allows control parameters to be set for emCgm.

USAGE:

emCgm.control(maxit = Inf, tolerance = 0.001, last = 2, trace = T) 

OPTIONAL ARGUMENTS:

maxit
the maximum number of iterations in the EM algorithm. Alternatively, maxit can be a list containing some or all parameters in the list returned by emCgm.control.
tolerance
convergence criterion for the EM algorithm. This may either be a number or a vector of two numbers. By default, convergence is assumed when the maximum relative change in the cell probability estimates is less than tolerance[1]. You may also specify tolerance[2], in which case convergence also occurs when the absolute change in the log-likelihood from one iteration to the next is less than tolerance[2]. By default, the log-likelihood is not used in checking for convergence.
last
for the EM algorithm, the sequence of iteration numbers that are saved is determined as the final last iterations.
trace
if TRUE, additonal information is printed during the EM algorithm.

VALUE:

a list containing the control values described above.

SEE ALSO:

, , .

EXAMPLES:

emCgm.control(maxit = 200, tolerance = 0.0001, last = 50, trace = T)