For description of the algorithm see J.A.Nelder and R.Mead, A Simplex Method for Minimization, Computer Journal, Vol.7, pp.308-313.
NQ = i2 = NP*NP + 4*NP + 1
IER = Greater than 0 if optimum is achieved
IER = 4 Variance in simplex is lt ACC**2, optimum achieved.Note: NMSIMP convergence depends on function values only
IER = -1 Iteration limit exceeded
IER = -2 Function error in first FUNC call
IER = -3 Input error
IER = -4 Not enough scratch storage
IER =-10 Number of permissible restarts exceeded
IER =-11 Error in simplex shrinkage
IER =-12 Error building initial simplex
IER =-13 No stopping criterion enabled
SECTION 5.1 OPTIONS
Besides the general options the following are available. The option which will most likely be of interest to the user is STP1.
STP1 = Initial simplex step size.
Default = 1.D0
VAR = Output, variance of function values in the simplex at the optimum
KONVGE = Number of steps taken before convergence is checked Default value is NP + 2
NRST = Number of restarts permitted (when simplex collapses). Default value is 0
Return to the Beginning