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
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
|
\name{plot.stepmented}
\alias{plot.stepmented}
%- Also NEED an '\alias' for EACH other topic documented here.
\title{
Plot method for stepmented objects
}
\description{
Takes a fitted \code{stepmented} object returned by \code{stepmented()} and plots (or adds)
the fitted piecewise constant lines for the selected stepmented term.
}
\usage{
\method{plot}{stepmented}(x, term, add = FALSE, res = TRUE, conf.level=0, interc = TRUE, add.fx = FALSE,
psi.lines = TRUE, link=TRUE, const=NULL, res.col=grey(.15, alpha = .4),
surf=FALSE, zero.cor=TRUE, heurs=TRUE, shade=FALSE, se.type=c("cdf","abs","none"),
k=NULL, .vcov=NULL, leg="topleft", ...)
}
%- maybe also 'usage' for other objects documented here.
\arguments{
\item{x}{
a fitted \code{stepmented} object.
}
\item{term}{
the stepmented variable having the piece-wise constant relationship to be plotted.
If there is a single stepmented variable in the fitted model \code{x},
\code{term} can be omitted.
}
\item{add}{
when \code{TRUE} the fitted lines are added to the current device.
}
\item{res}{
when \code{TRUE} the fitted lines are plotted along with corresponding partial residuals.
}
\item{conf.level}{
the confidence level for the pointwise confidence intervals for the expected values.
}
\item{interc}{
if \code{TRUE} the computed components include the model intercept (if it exists).
}
\item{add.fx}{
logical. If TRUE and the object fit also includes an additional term for the same stepmented variable, the plot also portrays such `additional' term.
}
\item{psi.lines}{
if \code{TRUE} vertical lines corresponding to the estimated changepoints are also drawn
}
\item{link}{
if \code{FALSE} the fitted lines (and possibily the residuals) are reported on the response scale. Ignored if the fit object \code{x} is not a glm-like fit.
}
\item{const}{ constant to add to each fitted segmented relationship (on the scale of the linear predictor) before plotting.
If \code{const=NULL} and the fit includes a segmented interaction term (obtained via \code{seg(..,by)} in the formula), the group-specific intercept is included.
}
\item{res.col}{when \code{res=TRUE} it means the color of the points representing the partial residuals.}
\item{surf}{
if the object fit \code{x} includes 2 stepmented covariates (x1 and x2, say) with relevant estimated breakpoints, \code{surf=TRUE} will draw on the plane x1-x2 the areas splitted according to the estimated breakpoints with corresponding estimated means superimposed.
}
\item{zero.cor}{
see \code{zero.cor} in \code{\link{vcov.stepmented}}; effective only if \code{conf.level>0}.
}
\item{heurs}{
logical; if \code{TRUE}, heuristic (usually somewhat conservative) confidence intervals are computed and plotted; effective only if \code{conf.level>0}.
}
\item{shade}{
if \code{TRUE} the pointwise confidence intervals are portrayed via shaded area; effective only if \code{conf.level>0}.
}
\item{se.type}{
which standard errors should be computed? see \code{type} in \code{\link{vcov.stepmented}}; effective only if \code{conf.level>0}.
}
\item{k}{
The value to be passed to \code{vcov.stepmented} to computed the standard errors.
}
\item{.vcov}{
The estimate var-covariance matrix; if \code{NULL}, it is computed internally by \code{\link{vcov.stepmented}}.
}
\item{leg}{
If the plot refers to stepmented relationships in groups, i.e. \code{term} has been specified as a vector, a legend is placed at the specified \code{leg} position. Put \code{NA} not to draw the legend.
}
\item{\dots}{ other graphics parameters to pass to plotting commands: `col', `lwd' and `lty' (that
can be vectors and are recycled if necessary, see the example below) for the fitted piecewise constant lines; `ylab', `xlab', `main', `sub', `cex.axis', `cex.lab', `xlim' and `ylim' when a new plot is produced (i.e. when \code{add=FALSE}); `pch' and `cex' for the partial residuals (when \code{res=TRUE}, \code{res.col} is for the color).% \code{col.shade} for the shaded regions (provided that \code{shade=TRUE} and \code{conf.level>0}).
}
}
\details{
Produces (or adds to the current device) the fitted step-function like relationship between the
response and the selected \code{term}. If the fitted model includes just a single `stepmented' variable,
\code{term} may be omitted. If \code{surf=TRUE}, and \code{res=TRUE} the point widths are proportional to the partial residual values.
}
\value{
None.
}
%\references{
%% ~put references to the literature/web site here ~
%}
\author{
Vito M. R. Muggeo
}
\note{
Implementation of confidence intervals for the conditional means in stepmented regression is under development; \code{conf.level>0} should be used with care, especially with multiple jumpoints.
}
%% ~Make other sections like Warning with \section{Warning }{....} ~
\seealso{
See Also as \code{\link{stepmented}}
}
\examples{
#Following code in stepmented..
\dontrun{
par(mfrow=c(1,3))
plot(os,"x")
plot(os,"z")
plot(os,"z", add.fx=TRUE, psi.lines=FALSE )
lines(os, "z")
#display the 'surface'
par(mfrow=c(1,3))
plot(os, surf=TRUE, col=1, res.col=2)
plot(os, surf=TRUE, lty=2)
plot(x,z)
plot(os, surf=TRUE, add=TRUE, col=4, res=FALSE)
}
}
\keyword{ regression }
\keyword{ nonlinear }
\keyword{ hplot }
% Use only one keyword per line.
% For non-standard keywords, use \concept instead of \keyword:
% \concept{ ~cpt1 }
% \concept{ ~cpt2 }
% Use only one concept per line.
|
