1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
|
\name{multistart}
\alias{multistart}
\encoding{UTF-8}
\title{General-purpose optimization - multiple starts}
\concept{minimization}
\concept{maximization}
\description{
Multiple initial parameter wrapper function that calls other
R tools for optimization, including the existing optimr() function.
}
\usage{
multistart(parmat, fn, gr=NULL, lower=-Inf, upper=Inf,
method=NULL, hessian=FALSE,
control=list(),
...)
}
\arguments{
\item{parmat}{a matrix of which each row is a set of initial values
for the parameters
for which optimal values are to be found. Names on the elements
of this vector are preserved and used in the results data frame.}
\item{fn}{A function to be minimized (or maximized), with first
argument the vector of parameters over which minimization is to take
place. It should return a scalar result.}
\item{gr}{A function to return (as a vector) the gradient for those methods that
can use this information.
If 'gr' is \code{NULL}, a finite-difference approximation will be used.
An open question concerns whether the SAME approximation code used for all methods,
or whether there are differences that could/should be examined? }
\item{lower, upper}{Bounds on the variables for methods such as \code{"L-BFGS-B"} that can
handle box (or bounds) constraints.}
\item{method}{A list of the methods to be used.
Note that this is an important change from optim() that allows
just one method to be specified. See \sQuote{Details}.
The default of NULL causes an appropriate set of methods to be supplied
depending on the presence or absence of bounds on the parameters. The default
unconstrained set is Rvmminu, Rcgminu, lbfgsb3, newuoa and nmkb.
The default bounds constrained set is Rvmminb, Rcgminb, lbfgsb3, bobyqa and nmkb.}
\item{hessian}{A logical control that if TRUE forces the computation of an approximation
to the Hessian at the final set of parameters. If FALSE (default), the hessian is
calculated if needed to provide the KKT optimality tests (see \code{kkt} in
\sQuote{Details} for the \code{control} list).
This setting is provided primarily for compatibility with optim().}
\item{control}{A list of control parameters. See \sQuote{Details}.}
\item{\dots}{For \code{optimx} further arguments to be passed to \code{fn}
and \code{gr}; otherwise, further arguments are not used.}
}
\details{
Note that arguments after \code{\dots} must be matched exactly.
See \code{optimr()} for other details.
}
\value{
An array with one row per set of starting parameters. Each row contains:
\item{par}{The best set of parameters found.}
\item{value}{The value of ‘fn’ corresponding to ‘par’.}
\item{counts}{ A two-element integer vector giving the number of calls to
‘fn’ and ‘gr’ respectively. This excludes those calls needed
to compute the Hessian, if requested, and any calls to ‘fn’
to compute a finite-difference approximation to the gradient.}
\item{convergence}{ An integer code. ‘0’ indicates successful completion}
\item{ message}{ A character string giving any additional information returned
by the optimizer, or ‘NULL’.}
\item{hessian}{ Always NULL for this routine.}
}
\source{
See the manual pages for \code{optim()} and the packages the DESCRIPTION \code{suggests}.
}
\examples{
fnR <- function (x, gs=100.0)
{
n <- length(x)
x1 <- x[2:n]
x2 <- x[1:(n - 1)]
sum(gs * (x1 - x2^2)^2 + (1 - x2)^2)
}
grR <- function (x, gs=100.0)
{
n <- length(x)
g <- rep(NA, n)
g[1] <- 2 * (x[1] - 1) + 4*gs * x[1] * (x[1]^2 - x[2])
if (n > 2) {
ii <- 2:(n - 1)
g[ii] <- 2 * (x[ii] - 1) + 4 * gs * x[ii] * (x[ii]^2 - x[ii +
1]) + 2 * gs * (x[ii] - x[ii - 1]^2)
}
g[n] <- 2 * gs * (x[n] - x[n - 1]^2)
g
}
pm <- rbind(rep(1,4), rep(pi, 4), rep(-2,4), rep(0,4), rep(20,4))
pm <- as.matrix(pm)
cat("multistart matrix:\n")
print(pm)
ans <- multistart(pm, fnR, grR, method="Rvmmin", control=list(trace=0))
ans
}
\keyword{nonlinear}
\keyword{optimize}
|
