Grocer Function

NAME

dfp_min - DFP minimization

CALLING SEQUENCE

[result]=dfp_min(func,pin[,arg1,...argn])

PARAMETERS

Input

Output

DESCRIPTION

DFP minimization routine to minimize func (Converted from Numerical Recipes book dfpmin routine). func must take the form func(b,varargin) where:
b = parameter vector (k x 1)
varargin = arguments passed to the function

EXAMPLE

1) res = ols2(y,x);
2) result4 = dfp_min('to_llike', [res('beta'); res('sige')],'maxit=1000',y,x);

This example, taken from optim1_d, realises the estimation of a tobit model, with initial values taken from on ols estimation. Maximum # of iterations is set to 1000.

AUTHOR

Eric Dubois 2002