Set Control Parameters for emGauss

DESCRIPTION:

Allows control parameters to be set for emGauss.

USAGE:

emGauss.control(maxit = Inf, tolerance = 0.001, 
    rcmin = sqrt(.Machine$double.eps), 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 emGauss.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 parameter 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.
rcmin
parameter used to determine when the variance-covariance matrix is singular. Typically, sqrt(.Machine$double.eps) is acceptable.
last
the sequence of iteration numbers that are saved is determined as the final last iterations.
trace
if TRUE, additional information is printed during the EM algorithm.

VALUE:

a list containing the control values described above.

SEE ALSO:

, , .

EXAMPLES:

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