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
107
108
109
110
111
112
113
114
115
116
117
|
\name{psi.findc}
\title{Find Tuning Constant(s) for "lqq" and "ggw" Psi Functions}
\alias{.psi.ggw.findc}% --> ../R/lmrob.MM.R
\alias{.psi.lqq.findc}
\alias{.psi.const}
\description{
Find psi function tuning constant sets for \code{"LQQ"} and \code{"GGW"}
psi (\eqn{\psi}) functions by specifying largest descent (minimal
slope), efficiency and or breakdown point.
\code{.psi.const()} is called from \code{\link{lmrob.control}()} to
set the tuning constants for psi and chi for \code{"LQQ"} and
\code{"GGW"} psi. Unless the specified tuning constants are from
fixed small set where the computations are stored precomputed,
\code{.psi.const()} calls the corresponding \code{.psi.<psi>.findc()}.
}
\usage{
.psi.ggw.findc(ms, b, eff = NA, bp = NA,
subdivisions = 100L,
rel.tol = .Machine$double.eps^0.25, abs.tol = rel.tol,
tol = .Machine$double.eps^0.25, ms.tol = tol/64, maxiter = 1000)
.psi.lqq.findc(ms, b.c, eff = NA, bp = NA,
interval = c(0.1, 4), subdivisions = 100L,
rel.tol = .Machine$double.eps^0.25, abs.tol = rel.tol,
tol = .Machine$double.eps^0.25, maxiter = 1000)
.psi.const(cc, psi)
}
\arguments{
\item{ms}{number, the minimal slope, typically negative.}
\item{b, b.c}{number, specifying \eqn{b} or \eqn{b/c} for \code{"ggw"}
or \code{"lqq"} respectively.}
\item{eff}{a number (or \code{\link{NA}}), the desired
\emph{efficiency}, in \eqn{[0,1]} of the estimator. If \code{NA},
\code{bp} must be specified as valid number.}
\item{bp}{a number (or \code{\link{NA}}), the desired
\emph{\bold{b}reakdown \bold{p}oint} of the estimator, in \eqn{[0,1]}.}
\item{interval}{for finding \eqn{c} via \code{\link{uniroot}()}.}
\item{subdivisions}{passed to \code{\link{integrate}()}.}
\item{rel.tol, abs.tol}{relative and absolute tolerance for
\code{\link{integrate}()}.}
\item{tol}{relative tolerance for \code{\link{uniroot}()}.}
\item{ms.tol}{relative tolerance for the internal
\code{.psi.ggw.finda()}, eventually passed to \code{\link{optimize}}
inside (internal) \code{.psi.ggw.mxs()}.}
\item{maxiter}{maximal number of iterations for
\code{\link{uniroot}()}.}
%% for .psi.const() :
\item{cc}{(for \code{.psi.const()}:) numeric vector of length 4,
containing all constants \code{c(ms, b*, eff, bp)}, where
\code{b* = b} for \code{"ggw"} and
\code{b* = b.c} for \code{"lqq"}, and one of \code{(eff, bp)}
is \code{NA}.}
\item{psi}{a string, either \code{"ggw"} or \code{"lqq"}.}
}
\details{
For some important special cases, the result of \code{.psi.*.findc()}
are stored precomputed for efficiency reasons. These cases are (the
defaults for \code{tuning.chi} and \code{tuning.psi} respectively in
\code{\link{lmrob.control}()}s result,
\tabular{ll}{
\code{tuning.chi} \tab\code{tuning.psi} \cr
\code{c(-0.5, 1.5, NA, 0.5)} \tab\code{c(-0.5, 1.5, 0.95, NA)}
}
and for \code{"ggw"} additionally, these four cases:
\tabular{ll}{
\code{tuning.chi} \tab\code{tuning.psi} \cr
{} \tab\code{c(-0.5, 1.5, 0.85, NA)} \cr
\code{c(-0.5, 1, NA, 0.5)} \tab\code{c(-0.5, 1, 0.95, NA)} \cr
{} \tab\code{c(-0.5, 1, 0.85, NA)}
} %tab
Note that for \code{"ggw"}, exactly these \eqn{2+4 = 6} cases also
allow fast \eqn{\rho} and \eqn{\chi} (aka
\eqn{\tilde\rho(\cdot)}{rho~(.)}, see \code{\link{Mchi}}),
function evaluations. For all other tuning constant settings, rho()
evaluations are based on numerical integration via \R's own
\code{Rdqags()} C function (part of R's official API).
}
\value{
a \code{\link{numeric}} vector of constants, for \code{"lqq"} or
\code{"ggw"} psi functions, respectively:
\describe{
\item{\code{"lqq"}:}{\eqn{(b, c, s) = (b/c * c, c, s = 1 - min_slope)},}
\item{\code{"ggw"}:}{\eqn{(0, a, b, c, \rho(\infty))}{%
(0, a, b, c, rho(Inf))}.}
}
\code{.psi.const(cc, psi)} returns the argument \code{cc} with the
above constant vectors as attribute \code{"constants"}, in the case of
\code{psi = "lqq"} in all cases (since \pkg{robustbase} version >=
0.93), for \code{psi = "ggw"} only in the non-standard cases.
}
\references{
See the vignette about %% ../vignettes/psi_functions.Rnw :
\dQuote{\eqn{\psi}{psi}-Functions Available in Robustbase}.
}
\seealso{\code{\link{Mpsi}()} etc for the psi function definitions;
\code{\link{.Mpsi.tuning.defaults}}, etc, for tuning constants'
defaults for \code{\link{lmrob}()}.
}
\author{Manuel Koller (original) and Martin Maechler (arguments, export, docs).
}
\examples{
(c.ge95 <- .psi.ggw.findc(ms = -0.5, b = 1.5, eff = 0.95))
(c.ge90 <- .psi.ggw.findc(ms = -0.5, b = 1.5, eff = 0.90))
(c.gb50 <- .psi.ggw.findc(ms = -0.5, b = 1.5, bp = 0.50))
stopifnot(all.equal(c.ge95, c(0, 1.386362, 1.5, 1.0628199, 4.7773893), tol = 1e-5),
all.equal(c.ge90, c(0, 1.0282811, 1.5, 0.87086259, 3.2075233), tol = 1e-5),
all.equal(c.gb50, c(0, 0.20367394, 1.5, 0.29591308, 0.37033962),tol = 1e-5))
(cl.e.95 <- .psi.lqq.findc(ms = -0.5, b.c = 1.5, eff = .95))
(cl.b.50 <- .psi.lqq.findc(ms = -0.5, b.c = 1.5, bp = .50))
stopifnot(all.equal(cl.e.95, c(1.4734061, 0.98227073, 1.5), tol = 1e-5),
all.equal(cl.b.50, c(0.40154568, 0.26769712, 1.5), tol = 1e-5))
}
\keyword{utilities}
|
