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
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
|
% Generated by roxygen2: do not edit by hand
% Please edit documentation in R/ppd-distributions.R
\name{PPD-distributions}
\alias{PPD-distributions}
\alias{ppd_data}
\alias{ppd_dens_overlay}
\alias{ppd_ecdf_overlay}
\alias{ppd_dens}
\alias{ppd_hist}
\alias{ppd_dots}
\alias{ppd_freqpoly}
\alias{ppd_freqpoly_grouped}
\alias{ppd_boxplot}
\title{PPD distributions}
\usage{
ppd_data(ypred, group = NULL)
ppd_dens_overlay(
ypred,
...,
size = 0.25,
alpha = 0.7,
trim = FALSE,
bw = "nrd0",
adjust = 1,
kernel = "gaussian",
n_dens = 1024
)
ppd_ecdf_overlay(
ypred,
...,
discrete = FALSE,
pad = TRUE,
size = 0.25,
alpha = 0.7
)
ppd_dens(ypred, ..., trim = FALSE, size = 0.5, alpha = 1)
ppd_hist(ypred, ..., binwidth = NULL, bins = NULL, breaks = NULL, freq = TRUE)
ppd_dots(ypred, ..., binwidth = NA, quantiles = NA, freq = TRUE)
ppd_freqpoly(
ypred,
...,
binwidth = NULL,
bins = NULL,
freq = TRUE,
size = 0.5,
alpha = 1
)
ppd_freqpoly_grouped(
ypred,
group,
...,
binwidth = NULL,
bins = NULL,
freq = TRUE,
size = 0.5,
alpha = 1
)
ppd_boxplot(ypred, ..., notch = TRUE, size = 0.5, alpha = 1)
}
\arguments{
\item{ypred}{An \code{S} by \code{N} matrix of draws from the posterior (or prior)
predictive distribution. The number of rows, \code{S}, is the size of the
posterior (or prior) sample used to generate \code{ypred}. The number of
columns, \code{N}, is the number of predicted observations.}
\item{group}{A grouping variable of the same length as \code{y}.
Will be coerced to \link[base:factor]{factor} if not already a factor.
Each value in \code{group} is interpreted as the group level pertaining
to the corresponding observation.}
\item{...}{For dot plots, optional additional arguments to pass to \code{\link[ggdist:stat_dots]{ggdist::stat_dots()}}.}
\item{size, alpha}{Passed to the appropriate geom to control the appearance of
the predictive distributions.}
\item{trim}{A logical scalar passed to \code{\link[ggplot2:geom_density]{ggplot2::geom_density()}}.}
\item{bw, adjust, kernel, n_dens}{Optional arguments passed to
\code{\link[stats:density]{stats::density()}} to override default kernel density estimation
parameters. \code{n_dens} defaults to \code{1024}.}
\item{discrete}{For \code{ppc_ecdf_overlay()}, should the data be treated as
discrete? The default is \code{FALSE}, in which case \code{geom="line"} is
passed to \code{\link[ggplot2:stat_ecdf]{ggplot2::stat_ecdf()}}. If \code{discrete} is set to
\code{TRUE} then \code{geom="step"} is used.}
\item{pad}{A logical scalar passed to \code{\link[ggplot2:stat_ecdf]{ggplot2::stat_ecdf()}}.}
\item{binwidth}{Passed to \code{\link[ggplot2:geom_histogram]{ggplot2::geom_histogram()}} to override
the default binwidth.}
\item{bins}{Passed to \code{\link[ggplot2:geom_histogram]{ggplot2::geom_histogram()}} to override
the default binwidth.}
\item{breaks}{Passed to \code{\link[ggplot2:geom_histogram]{ggplot2::geom_histogram()}} as an
alternative to \code{binwidth}.}
\item{freq}{For histograms, \code{freq=TRUE} (the default) puts count on the
y-axis. Setting \code{freq=FALSE} puts density on the y-axis. (For many
plots the y-axis text is off by default. To view the count or density
labels on the y-axis see the \code{\link[=yaxis_text]{yaxis_text()}} convenience
function.)}
\item{quantiles}{For dot plots, an optional integer passed to
\code{\link[ggdist:stat_dots]{ggdist::stat_dots()}} specifying the number of quantiles to use for a
quantile dot plot. If \code{quantiles} is \code{NA} (the default) then all data
points are plotted.}
\item{notch}{For the box plot, a logical scalar passed to
\code{\link[ggplot2:geom_boxplot]{ggplot2::geom_boxplot()}}. Note: unlike \code{geom_boxplot()}, the default is
\code{notch=TRUE}.}
}
\value{
The plotting functions return a ggplot object that can be further
customized using the \strong{ggplot2} package. The functions with suffix
\verb{_data()} return the data that would have been drawn by the plotting
function.
}
\description{
Plot posterior or prior predictive distributions. Each of these functions
makes the same plot as the corresponding \code{\link[=PPC-distributions]{ppc_}} function
but without plotting any observed data \code{y}. The \strong{Plot Descriptions} section
at \link{PPC-distributions} has details on the individual plots.
}
\details{
For Binomial data, the plots may be more useful if
the input contains the "success" \emph{proportions} (not discrete
"success" or "failure" counts).
}
\examples{
# difference between ppd_dens_overlay() and ppc_dens_overlay()
color_scheme_set("brightblue")
preds <- example_yrep_draws()
ppd_dens_overlay(ypred = preds[1:50, ])
ppc_dens_overlay(y = example_y_data(), yrep = preds[1:50, ])
}
\seealso{
Other PPDs:
\code{\link{PPD-intervals}},
\code{\link{PPD-overview}},
\code{\link{PPD-test-statistics}}
}
\concept{PPDs}
|
