fit.control {BTSR}R Documentation

Default control list

Description

Sets default values for constants used by the optimization functions in FORTRAN.

Usage

fit.control(control = list())

Arguments

control

a list with configurations to be passed to the optimization subroutines. Missing arguments will receive default values. See ‘Details’.

Details

The control argument is a list that can supply any of the following components

method

The optimization method. Current available options are "L-BFGS-B" and "Nelder-Mead". Default is "L-BFGS-B".

maxit

The maximum number of iterations. Defaults is 1000.

iprint

The frequency of reports if control$trace is positive. Defaults is -1 (no report).

  • For "L-BFGS-B" method:

    • iprint < 0 no output is generated;

    • iprint = 0 print only one line at the last iteration;

    • ⁠0 < iprint < 99⁠ print also f and |proj g| every iprint iterations;

    • iprint = 99 print details of every iteration except n-vectors;

    • iprint = 100 print also the changes of active set and final x;

    • iprint > 100 print details of every iteration including x and g;

  • For "Nelder-Mead" method:

    • iprint < 0 No printing

    • iprint = 0 Printing of parameter values and the function value after initial evidence of convergence.

    • iprint > 0 As for iprint = 0 plus progress reports after every iprint evaluations, plus printing for the initial simplex.

factr

controls the convergence of the "L-BFGS-B" method. Convergence occurs when the reduction in the objective is within this factor of the machine tolerance. The iteration will stop when

\dfrac{(f^k - f^{k+1})}{\max\{|f^k|,|f^{k+1}|,1\}} \le factr \times epsmch

where epsmch is the machine precision, which is automatically generated by the code. Typical values for factr: 1.e+12 for low accuracy; 1.e+7 for moderate accuracy; 1.e+1 for extremely high accuracy. Default is 1e7, that is a tolerance of about 1e-8.

pgtol

helps control the convergence of the "L-BFGS-B" method. It is a tolerance on the projected gradient in the current search direction. the iteration will stop when

\max\{|\text{proj }g_i |, i = 1, ..., n\} \le pgtol

where \text{proj }g_i is the ith component of the projected gradient. Default is 1e-12.

stopcr

The criterion applied to the standard deviation of the values of objective function at the points of the simplex, for "Nelder-Mead" method.

Value

Returns a list with all arguments in ‘Details’.

Examples

BTSR::fit.control()


[Package BTSR version 1.0.0 Index]