Nothing
R/PLOT.R
In UPG: Efficient Bayesian Algorithms for Binary and Categorical Data Regression Models
Defines functions
plot.UPG.Binomial
plot.UPG.MNL
plot.UPG.Logit
plot.UPG.Probit
Documented in
plot.UPG.Binomial
plot.UPG.Logit
plot.UPG.MNL
plot.UPG.Probit
#' @name plot.UPG.Probit
#' @title Coefficient plots for UPG.Probit objects
#'
#' @description \code{plot} generates plots from \code{UPG.Probit} objects using \code{ggplot2}. Coefficient plots show point estimates for all coefficients as well as their credible intervals.
#'
#' @param x an object of class \code{UPG.Probit}.
#' @param sort a logical variable indicating whether the plotted coefficients should be sorted according to effect sizes. Default is FALSE.
#' @param names a character vector indicating names for the variables used in the plots.
#' @param xlab a character vector of length 1 indicating a title for the x-axis.
#' @param ylab a character vector of length 1 indicating a title for the y-axis.
#' @param q a numerical vector of length two providing the posterior quantiles to be extracted. Default are 0.025 and 0.975 quantiles.
#' @param include can be used to plot only a subset of variables. Specify the columns of X that should be kept in the plot. See examples for further information.
#' @param ... other plot parameters.
#'
#' @return Returns a ggplot2 object.
#'
#' @seealso
#' \code{\link{summary.UPG.Probit}} to summarize a \code{UPG.Probit} object and create tables.
#' \code{\link{predict.UPG.Probit}} to predict probabilities using a \code{UPG.Probit} object.
#' \code{\link{coef.UPG.Probit}} to extract coefficients from a \code{UPG.Probit} object.
#'
#' @author Gregor Zens
#'
#' @examples
#' \donttest{
#' # estimate a probit model using example data
#' library(UPG)
#' data(lfp)
#' y = lfp[,1]
#' X = lfp[,-1]
#' results.probit = UPG(y = y, X = X, model = "probit")
#'
#' # plot the results and sort coefficients by effect size
#' plot(results.probit, sort = TRUE)
#'
#' # plot only variables 1 and 3 with custom names, credible intervals and axis labels
#' plot(results.probit,
#' include = c(1, 3),
#' names = c("Custom 1", "Custom 2"),
#' q = c(0.1, 0.9),
#' xlab = c("Custom X"),
#' ylab = c("Custom Y"))
#'}
#' @method plot UPG.Probit
#'
#'@export
plot.UPG.Probit = function(x = NULL,
...,
sort = FALSE, # sort coefficients by average effect size
names = NULL, # provide names for variables, alternatively
xlab = NULL, # provide x axis label
ylab = NULL, # provide y axis label
q = c(0.025, 0.975), # credible intervals
include = NULL # which variables to include? default:all (numeric vector)
){
if(is.null(include)) include = 1:ncol(x$inputs$X)
if(is.null(names)) names = colnames(x$inputs$X[,include,drop=F])
if(is.null(names)) names = paste0("Variable", 1:ncol(x$inputs$X[,include,drop=F]))
if(length(names) != length(include)) stop("Number of provided variable names does not match number of included variables.")
#create some global variables such that r cmd check is happy
variable = iter = value = NULL
c.point = apply(x$posterior$beta[,include,drop=F], 2, mean)
c.upper = apply(x$posterior$beta[,include,drop=F], 2, quantile, q[2])
c.lower = apply(x$posterior$beta[,include,drop=F], 2, quantile, q[1])
plot.df = data.frame(c.point,c.upper,c.lower,names)
if(nrow(plot.df) > 1 & sort){
plot.df$names = factor(plot.df$names, levels = plot.df$names[order(plot.df$c.point)])
}
if(!sort){
# plot in order of appearance in X
plot.df$names = factor(plot.df$names, levels = rev(plot.df$names))
}
#axis labeling, be careful b/c of coord_flip
if(is.null(ylab)) ylab = ""
if(is.null(xlab)) xlab = "Posterior Estimate"
final = ggplot(plot.df, aes(x=names, y=c.point)) +
geom_errorbar(ymin=c.lower, ymax=c.upper, width=0, col="grey60") +
geom_point() +
geom_point(data = plot.df, aes(x=names, y=c.upper), size=-1) + #for correct scaling
geom_point(data = plot.df, aes(x=names, y=c.lower), size=-1) + #for correct scaling
theme_bw() +
geom_hline(yintercept = 0,lty=2) +
coord_flip() +
xlab(ylab) +
ylab(xlab)
return(final)
}
#' @name plot.UPG.Logit
#' @title Coefficient plots for UPG.Logit objects
#'
#' @description \code{plot} generates plots from \code{UPG.Logit} objects using \code{ggplot2}. Coefficient plots show point estimates for all coefficients as well as their credible intervals.
#'
#' @param x an object of class \code{UPG.Logit}.
#' @param sort a logical variable indicating whether the plotted coefficients should be sorted according to effect sizes. Default is FALSE.
#' @param names a character vector indicating names for the variables used in the plots.
#' @param xlab a character vector of length 1 indicating a title for the x-axis.
#' @param ylab a character vector of length 1 indicating a title for the y-axis.
#' @param q a numerical vector of length two providing the posterior quantiles to be extracted. Default are 0.025 and 0.975 quantiles.
#' @param include can be used to plot only a subset of variables. Specify the columns of X that should be kept in the plot. See examples for further information.
#' @param ... other plot parameters.
#'
#' @return Returns a ggplot2 object.
#'
#' @seealso
#' \code{\link{summary.UPG.Logit}} to summarize a \code{UPG.Logit} object and create tables.
#' \code{\link{predict.UPG.Logit}} to predict probabilities using a \code{UPG.Logit} object.
#' \code{\link{coef.UPG.Logit}} to extract coefficients from a \code{UPG.Logit} object.
#'
#' @author Gregor Zens
#'
#' @examples
#' \donttest{
#' # estimate a logit model using example data
#' library(UPG)
#' data(lfp)
#' y = lfp[,1]
#' X = lfp[,-1]
#' results.logit = UPG(y = y, X = X, model = "logit")
#'
#' # plot the results and sort coefficients by effect size
#' plot(results.logit, sort = TRUE)
#'
#' # plot only variables 1 and 3 with custom names, credible intervals and axis labels
#' plot(results.logit,
#' include = c(1,3),
#' names = c("Custom 1", "Custom 2"),
#' q = c(0.1, 0.9),
#' xlab = c("Custom X"),
#' ylab = c("Custom Y"))
#'}
#' @method plot UPG.Logit
#'
#'@export
plot.UPG.Logit = function(x = NULL,
...,
sort = FALSE, # sort coefficients by average effect size
names = NULL, # provide names for variables, alternatively
xlab = NULL, # provide x axis label
ylab = NULL, # provide y axis label
q = c(0.025, 0.975), # credible intervals
include = NULL # which variables to include? default:all (numeric vector)
){
if(is.null(include)) include = 1:ncol(x$inputs$X)
if(is.null(names)) names = colnames(x$inputs$X[,include,drop=F])
if(is.null(names)) names = paste0("Variable", 1:ncol(x$inputs$X[,include,drop=F]))
if(length(names) != length(include)) stop("Number of provided variable names does not match number of included variables.")
#create some global variables such that r cmd check is happy
variable = iter = value = NULL
c.point = apply(x$posterior$beta[,include,drop=F], 2, mean)
c.upper = apply(x$posterior$beta[,include,drop=F], 2, quantile, q[2])
c.lower = apply(x$posterior$beta[,include,drop=F], 2, quantile, q[1])
plot.df = data.frame(c.point,c.upper,c.lower,names)
if(nrow(plot.df) > 1 & sort){
plot.df$names = factor(plot.df$names, levels = plot.df$names[order(plot.df$c.point)])
}
if(!sort){
# plot in order of appearance in X
plot.df$names = factor(plot.df$names, levels = rev(plot.df$names))
}
#axis labeling, be careful b/c of coord_flip
if(is.null(ylab)) ylab = ""
if(is.null(xlab)) xlab = "Posterior Estimate"
final = ggplot(plot.df, aes(x=names, y=c.point)) +
geom_errorbar(ymin=c.lower, ymax=c.upper, width=0, col="grey60") +
geom_point() +
geom_point(data = plot.df, aes(x=names, y=c.upper), size=-1) + #for correct scaling
geom_point(data = plot.df, aes(x=names, y=c.lower), size=-1) + #for correct scaling
theme_bw() +
geom_hline(yintercept = 0,lty=2) +
coord_flip() +
xlab(ylab) +
ylab(xlab)
return(final)
}
#' @name plot.UPG.MNL
#' @title Coefficient plots for UPG.MNL objects
#'
#' @description \code{plot} generates plots from \code{UPG.MNL} objects using \code{ggplot2}. Coefficient plots show point estimates for all coefficients in all groups except the baseline as well as their credible intervals.
#'
#' @param x an object of class \code{UPG.MNL}.
#' @param sort a logical variable indicating whether the plotted coefficients should be sorted according to average effect sizes across groups. Default is FALSE.
#' @param names a character vector indicating names for the variables used in the plots.
#' @param groups a character vector indicating names for the groups excluding the baseline. The group names must correspond to the ordering in the dependent variable used for estimation.
#' @param xlab a character vector of length 1 indicating a title for the x-axis.
#' @param ylab a character vector of length 1 indicating a title for the y-axis.
#' @param q a numerical vector of length two providing the posterior quantiles to be extracted. Default are 0.025 and 0.975 quantiles.
#' @param include can be used to plot only a subset of variables. Specify the columns of X that should be kept in the plot. See examples for further information.
#' @param ... other plot parameters.
#'
#' @return Returns a ggplot2 object.
#'
#' @seealso
#' \code{\link{summary.UPG.MNL}} to summarize a \code{UPG.MNL} object and create tables.
#' \code{\link{predict.UPG.MNL}} to predict probabilities using a \code{UPG.MNL} object.
#' \code{\link{coef.UPG.MNL}} to extract coefficients from a \code{UPG.MNL} object.
#'
#' @author Gregor Zens
#'
#' @examples
#' \donttest{
#' # estimate a multinomial logit model using example data
#' library(UPG)
#' data(program)
#' y = program[,1]
#' X = program[,-1]
#' results.mnl = UPG(y = y, X = X, model = "mnl")
#'
#' # plot the results and sort coefficients by average effect size
#' plot(results.mnl, sort = TRUE)
#'
#' # plot only variables 1 and 3 with custom group and variable names
#' # also, customize credible intervals and axis labels
#' plot(results.mnl,
#' include = c(1,3),
#' names = c("Custom 1", "Custom 2"),
#' groups = c("Alpha", "Beta"),
#' q = c(0.1, 0.9),
#' xlab = c("Custom X"),
#' ylab = c("Custom Y"))
#'}
#' @method plot UPG.MNL
#'
#'@export
plot.UPG.MNL = function(x = NULL,
...,
sort = FALSE, # sort coefficients by average effect size
names = NULL, # provide names for variables, alternatively
groups = NULL, # provide names for groups except baseline
xlab = NULL, # provide x axis label
ylab = NULL, # provide y axis label
q = c(0.025, 0.975), # credible intervals
include = NULL # which variables to include? default:all (numeric vector)
){
if(is.null(include)) include = 1:ncol(x$inputs$X)
if(is.null(names)) names = colnames(x$inputs$X[,include,drop=F])
if(is.null(names)) names = paste0("Variable", 1:ncol(x$inputs$X[,include,drop=F]))
if(length(names) != length(include)) stop("Number of provided variable names does not match number of included variables.")
#create some global variables such that r cmd check is happy
variable = iter = value = NULL
c.point = apply(x$posterior$beta[,include,,drop=F], c(2,3), mean)
c.upper = apply(x$posterior$beta[,include,,drop=F], c(2,3), quantile, q[2])
c.lower = apply(x$posterior$beta[,include,,drop=F], c(2,3), quantile, q[1])
if(nrow(c.point) == 1){c.upper = matrix(c.upper,nrow=1); c.lower = matrix(c.lower,nrow=1)}
if(is.null(groups)) groups = paste(x$posterior$groups[-ncol(c.point)])
if(length(groups) != ncol(c.point)-1) stop("Wrong number of group names supplied. Need K-1 names where K is the number of choices.")
#kick baseline
c.upper = c.upper[,-ncol(c.upper),drop=F]
c.lower = c.lower[,-ncol(c.lower),drop=F]
c.point = c.point[,-ncol(c.point),drop=F]
#add variable names
c.upper = data.frame(c.upper, names)
c.lower = data.frame(c.lower, names)
c.point = data.frame(c.point, names)
#add group names
colnames(c.upper) = colnames(c.lower) = colnames(c.point) = c(groups, "names")
#add measurement
c.upper$measure = "c.upper"
c.lower$measure = "c.lower"
c.point$measure = "c.point"
plot.df = rbind(c.upper,c.lower,c.point)
plot.df = reshape2::melt(plot.df, id.vars = c("names","measure"))
plot.df = reshape2::dcast(plot.df, "names + variable ~ measure")
#sorting (bit more complicated in MNL, I use average point estimate over all groups)
if(length(names)>1){
if(sort){
average = aggregate(plot.df$c.point, by=list(plot.df$names), FUN=mean)
lvls = unique(plot.df$names)
plot.df$names = factor(plot.df$names, levels = lvls[order(average$x)])
}
}
if(!sort){
# plot in order of appearance in X
plot.df$names = factor(plot.df$names, levels = rev(names))
}
#axis labeling, take care b/c of coord_flip
if(is.null(ylab)) ylab = ""
if(is.null(xlab)) xlab = "Posterior Estimate"
final = ggplot(plot.df, aes(x=names, y=c.point, shape = variable, group = variable)) +
geom_errorbar(aes(x=names, ymin=c.lower, ymax=c.upper, group = variable),
position = position_dodge(width=.5),
width=0, col="grey60") +
geom_point(position = position_dodge(width=.5)) +
theme_bw() +
geom_hline(yintercept = 0,lty=2) +
coord_flip() +
xlab(ylab) +
ylab(xlab) +
theme(legend.position = "bottom",
legend.title = element_blank())
return(final)
}
#' @name plot.UPG.Binomial
#' @title Coefficient plots for UPG.Binomial objects
#'
#' @description \code{plot} generates plots from \code{UPG.Binomial} objects using \code{ggplot2}. Coefficient plots show point estimates for all coefficients as well as their credible intervals.
#'
#' @param x an object of class \code{UPG.Binomial}.
#' @param sort a logical variable indicating whether the plotted coefficients should be sorted according to effect sizes. Default is FALSE.
#' @param names a character vector indicating names for the variables used in the plots.
#' @param xlab a character vector of length 1 indicating a title for the x-axis.
#' @param ylab a character vector of length 1 indicating a title for the y-axis.
#' @param q a numerical vector of length two providing the posterior quantiles to be extracted. Default are 0.025 and 0.975 quantiles.
#' @param include can be used to plot only a subset of variables. Specify the columns of X that should be kept in the plot. See examples for further information.
#' @param ... other plot parameters.
#'
#' @return Returns a ggplot2 object.
#'
#' @seealso
#' \code{\link{summary.UPG.Binomial}} to summarize a \code{UPG.Binomial} object and create tables.
#' \code{\link{predict.UPG.Binomial}} to predict probabilities using a \code{UPG.Binomial} object.
#' \code{\link{coef.UPG.Binomial}} to extract coefficients from a \code{UPG.Binomial} object.
#'
#' @author Gregor Zens
#'
#' @examples
#' \donttest{
#' # estimate a binomial logit model using example data
#' library(UPG)
#' data(titanic)
#' y = titanic[,1]
#' Ni = titanic[,2]
#' X = titanic[,-c(1,2)]
#' results.binomial = UPG(y = y, X = X, Ni = Ni, model = "binomial")
#'
#' # plot the results and sort coefficients by effect size
#' plot(results.binomial, sort = TRUE)
#'
#' # plot only variables 1 and 3 with custom names, credible intervals and axis labels
#' plot(results.binomial,
#' include = c(1,3),
#' names = c("Custom 1", "Custom 2"),
#' q = c(0.1, 0.9),
#' xlab = c("Custom X"),
#' ylab = c("Custom Y"))
#'}
#' @method plot UPG.Binomial
#'
#'@export
plot.UPG.Binomial = function(x = NULL,
...,
sort = FALSE, # sort coefficients by average effect size
names = NULL, # provide names for variables, alternatively
xlab = NULL, # provide x axis label
ylab = NULL, # provide y axis label
q = c(0.025, 0.975), # credible intervals
include = NULL # which variables to include? default:all (numeric vector)
){
if(is.null(include)) include = 1:ncol(x$inputs$X)
if(is.null(names)) names = colnames(x$inputs$X[,include,drop=F])
if(is.null(names)) names = paste0("Variable", 1:ncol(x$inputs$X[,include,drop=F]))
if(length(names) != length(include)) stop("Number of provided variable names does not match number of included variables.")
#create some global variables such that r cmd check is happy
variable = iter = value = NULL
c.point = apply(x$posterior$beta[,include,drop=F], 2, mean)
c.upper = apply(x$posterior$beta[,include,drop=F], 2, quantile, q[2])
c.lower = apply(x$posterior$beta[,include,drop=F], 2, quantile, q[1])
plot.df = data.frame(c.point,c.upper,c.lower,names)
if(nrow(plot.df) > 1 & sort){
plot.df$names = factor(plot.df$names, levels = plot.df$names[order(plot.df$c.point)])
}
if(!sort){
# plot in order of appearance in X
plot.df$names = factor(plot.df$names, levels = rev(plot.df$names))
}
#axis labeling, be careful b/c of coord_flip
if(is.null(ylab)) ylab = ""
if(is.null(xlab)) xlab = "Posterior Estimate"
final = ggplot(plot.df, aes(x=names, y=c.point)) +
geom_errorbar(ymin=c.lower, ymax=c.upper, width=0, col="grey60") +
geom_point() +
geom_point(data = plot.df, aes(x=names, y=c.upper), size=-1) + #for correct scaling
geom_point(data = plot.df, aes(x=names, y=c.lower), size=-1) + #for correct scaling
theme_bw() +
geom_hline(yintercept = 0,lty=2) +
coord_flip() +
xlab(ylab) +
ylab(xlab)
return(final)
}
Try the UPG package in your browser
Any scripts or data that you put into this service are public.
UPG documentation built on Nov. 4, 2023, 5:06 p.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Defines functions plot.UPG.Binomial plot.UPG.MNL plot.UPG.Logit plot.UPG.Probit
Documented in plot.UPG.Binomial plot.UPG.Logit plot.UPG.MNL plot.UPG.Probit
#' @name plot.UPG.Probit
#' @title Coefficient plots for UPG.Probit objects
#'
#' @description \code{plot} generates plots from \code{UPG.Probit} objects using \code{ggplot2}. Coefficient plots show point estimates for all coefficients as well as their credible intervals.
#'
#' @param x an object of class \code{UPG.Probit}.
#' @param sort a logical variable indicating whether the plotted coefficients should be sorted according to effect sizes. Default is FALSE.
#' @param names a character vector indicating names for the variables used in the plots.
#' @param xlab a character vector of length 1 indicating a title for the x-axis.
#' @param ylab a character vector of length 1 indicating a title for the y-axis.
#' @param q a numerical vector of length two providing the posterior quantiles to be extracted. Default are 0.025 and 0.975 quantiles.
#' @param include can be used to plot only a subset of variables. Specify the columns of X that should be kept in the plot. See examples for further information.
#' @param ... other plot parameters.
#'
#' @return Returns a ggplot2 object.
#'
#' @seealso
#' \code{\link{summary.UPG.Probit}} to summarize a \code{UPG.Probit} object and create tables.
#' \code{\link{predict.UPG.Probit}} to predict probabilities using a \code{UPG.Probit} object.
#' \code{\link{coef.UPG.Probit}} to extract coefficients from a \code{UPG.Probit} object.
#'
#' @author Gregor Zens
#'
#' @examples
#' \donttest{
#' # estimate a probit model using example data
#' library(UPG)
#' data(lfp)
#' y = lfp[,1]
#' X = lfp[,-1]
#' results.probit = UPG(y = y, X = X, model = "probit")
#'
#' # plot the results and sort coefficients by effect size
#' plot(results.probit, sort = TRUE)
#'
#' # plot only variables 1 and 3 with custom names, credible intervals and axis labels
#' plot(results.probit,
#' include = c(1, 3),
#' names = c("Custom 1", "Custom 2"),
#' q = c(0.1, 0.9),
#' xlab = c("Custom X"),
#' ylab = c("Custom Y"))
#'}
#' @method plot UPG.Probit
#'
#'@export
plot.UPG.Probit = function(x = NULL,
...,
sort = FALSE, # sort coefficients by average effect size
names = NULL, # provide names for variables, alternatively
xlab = NULL, # provide x axis label
ylab = NULL, # provide y axis label
q = c(0.025, 0.975), # credible intervals
include = NULL # which variables to include? default:all (numeric vector)
){
if(is.null(include)) include = 1:ncol(x$inputs$X)
if(is.null(names)) names = colnames(x$inputs$X[,include,drop=F])
if(is.null(names)) names = paste0("Variable", 1:ncol(x$inputs$X[,include,drop=F]))
if(length(names) != length(include)) stop("Number of provided variable names does not match number of included variables.")
#create some global variables such that r cmd check is happy
variable = iter = value = NULL
c.point = apply(x$posterior$beta[,include,drop=F], 2, mean)
c.upper = apply(x$posterior$beta[,include,drop=F], 2, quantile, q[2])
c.lower = apply(x$posterior$beta[,include,drop=F], 2, quantile, q[1])
plot.df = data.frame(c.point,c.upper,c.lower,names)
if(nrow(plot.df) > 1 & sort){
plot.df$names = factor(plot.df$names, levels = plot.df$names[order(plot.df$c.point)])
}
if(!sort){
# plot in order of appearance in X
plot.df$names = factor(plot.df$names, levels = rev(plot.df$names))
}
#axis labeling, be careful b/c of coord_flip
if(is.null(ylab)) ylab = ""
if(is.null(xlab)) xlab = "Posterior Estimate"
final = ggplot(plot.df, aes(x=names, y=c.point)) +
geom_errorbar(ymin=c.lower, ymax=c.upper, width=0, col="grey60") +
geom_point() +
geom_point(data = plot.df, aes(x=names, y=c.upper), size=-1) + #for correct scaling
geom_point(data = plot.df, aes(x=names, y=c.lower), size=-1) + #for correct scaling
theme_bw() +
geom_hline(yintercept = 0,lty=2) +
coord_flip() +
xlab(ylab) +
ylab(xlab)
return(final)
}
#' @name plot.UPG.Logit
#' @title Coefficient plots for UPG.Logit objects
#'
#' @description \code{plot} generates plots from \code{UPG.Logit} objects using \code{ggplot2}. Coefficient plots show point estimates for all coefficients as well as their credible intervals.
#'
#' @param x an object of class \code{UPG.Logit}.
#' @param sort a logical variable indicating whether the plotted coefficients should be sorted according to effect sizes. Default is FALSE.
#' @param names a character vector indicating names for the variables used in the plots.
#' @param xlab a character vector of length 1 indicating a title for the x-axis.
#' @param ylab a character vector of length 1 indicating a title for the y-axis.
#' @param q a numerical vector of length two providing the posterior quantiles to be extracted. Default are 0.025 and 0.975 quantiles.
#' @param include can be used to plot only a subset of variables. Specify the columns of X that should be kept in the plot. See examples for further information.
#' @param ... other plot parameters.
#'
#' @return Returns a ggplot2 object.
#'
#' @seealso
#' \code{\link{summary.UPG.Logit}} to summarize a \code{UPG.Logit} object and create tables.
#' \code{\link{predict.UPG.Logit}} to predict probabilities using a \code{UPG.Logit} object.
#' \code{\link{coef.UPG.Logit}} to extract coefficients from a \code{UPG.Logit} object.
#'
#' @author Gregor Zens
#'
#' @examples
#' \donttest{
#' # estimate a logit model using example data
#' library(UPG)
#' data(lfp)
#' y = lfp[,1]
#' X = lfp[,-1]
#' results.logit = UPG(y = y, X = X, model = "logit")
#'
#' # plot the results and sort coefficients by effect size
#' plot(results.logit, sort = TRUE)
#'
#' # plot only variables 1 and 3 with custom names, credible intervals and axis labels
#' plot(results.logit,
#' include = c(1,3),
#' names = c("Custom 1", "Custom 2"),
#' q = c(0.1, 0.9),
#' xlab = c("Custom X"),
#' ylab = c("Custom Y"))
#'}
#' @method plot UPG.Logit
#'
#'@export
plot.UPG.Logit = function(x = NULL,
...,
sort = FALSE, # sort coefficients by average effect size
names = NULL, # provide names for variables, alternatively
xlab = NULL, # provide x axis label
ylab = NULL, # provide y axis label
q = c(0.025, 0.975), # credible intervals
include = NULL # which variables to include? default:all (numeric vector)
){
if(is.null(include)) include = 1:ncol(x$inputs$X)
if(is.null(names)) names = colnames(x$inputs$X[,include,drop=F])
if(is.null(names)) names = paste0("Variable", 1:ncol(x$inputs$X[,include,drop=F]))
if(length(names) != length(include)) stop("Number of provided variable names does not match number of included variables.")
#create some global variables such that r cmd check is happy
variable = iter = value = NULL
c.point = apply(x$posterior$beta[,include,drop=F], 2, mean)
c.upper = apply(x$posterior$beta[,include,drop=F], 2, quantile, q[2])
c.lower = apply(x$posterior$beta[,include,drop=F], 2, quantile, q[1])
plot.df = data.frame(c.point,c.upper,c.lower,names)
if(nrow(plot.df) > 1 & sort){
plot.df$names = factor(plot.df$names, levels = plot.df$names[order(plot.df$c.point)])
}
if(!sort){
# plot in order of appearance in X
plot.df$names = factor(plot.df$names, levels = rev(plot.df$names))
}
#axis labeling, be careful b/c of coord_flip
if(is.null(ylab)) ylab = ""
if(is.null(xlab)) xlab = "Posterior Estimate"
final = ggplot(plot.df, aes(x=names, y=c.point)) +
geom_errorbar(ymin=c.lower, ymax=c.upper, width=0, col="grey60") +
geom_point() +
geom_point(data = plot.df, aes(x=names, y=c.upper), size=-1) + #for correct scaling
geom_point(data = plot.df, aes(x=names, y=c.lower), size=-1) + #for correct scaling
theme_bw() +
geom_hline(yintercept = 0,lty=2) +
coord_flip() +
xlab(ylab) +
ylab(xlab)
return(final)
}
#' @name plot.UPG.MNL
#' @title Coefficient plots for UPG.MNL objects
#'
#' @description \code{plot} generates plots from \code{UPG.MNL} objects using \code{ggplot2}. Coefficient plots show point estimates for all coefficients in all groups except the baseline as well as their credible intervals.
#'
#' @param x an object of class \code{UPG.MNL}.
#' @param sort a logical variable indicating whether the plotted coefficients should be sorted according to average effect sizes across groups. Default is FALSE.
#' @param names a character vector indicating names for the variables used in the plots.
#' @param groups a character vector indicating names for the groups excluding the baseline. The group names must correspond to the ordering in the dependent variable used for estimation.
#' @param xlab a character vector of length 1 indicating a title for the x-axis.
#' @param ylab a character vector of length 1 indicating a title for the y-axis.
#' @param q a numerical vector of length two providing the posterior quantiles to be extracted. Default are 0.025 and 0.975 quantiles.
#' @param include can be used to plot only a subset of variables. Specify the columns of X that should be kept in the plot. See examples for further information.
#' @param ... other plot parameters.
#'
#' @return Returns a ggplot2 object.
#'
#' @seealso
#' \code{\link{summary.UPG.MNL}} to summarize a \code{UPG.MNL} object and create tables.
#' \code{\link{predict.UPG.MNL}} to predict probabilities using a \code{UPG.MNL} object.
#' \code{\link{coef.UPG.MNL}} to extract coefficients from a \code{UPG.MNL} object.
#'
#' @author Gregor Zens
#'
#' @examples
#' \donttest{
#' # estimate a multinomial logit model using example data
#' library(UPG)
#' data(program)
#' y = program[,1]
#' X = program[,-1]
#' results.mnl = UPG(y = y, X = X, model = "mnl")
#'
#' # plot the results and sort coefficients by average effect size
#' plot(results.mnl, sort = TRUE)
#'
#' # plot only variables 1 and 3 with custom group and variable names
#' # also, customize credible intervals and axis labels
#' plot(results.mnl,
#' include = c(1,3),
#' names = c("Custom 1", "Custom 2"),
#' groups = c("Alpha", "Beta"),
#' q = c(0.1, 0.9),
#' xlab = c("Custom X"),
#' ylab = c("Custom Y"))
#'}
#' @method plot UPG.MNL
#'
#'@export
plot.UPG.MNL = function(x = NULL,
...,
sort = FALSE, # sort coefficients by average effect size
names = NULL, # provide names for variables, alternatively
groups = NULL, # provide names for groups except baseline
xlab = NULL, # provide x axis label
ylab = NULL, # provide y axis label
q = c(0.025, 0.975), # credible intervals
include = NULL # which variables to include? default:all (numeric vector)
){
if(is.null(include)) include = 1:ncol(x$inputs$X)
if(is.null(names)) names = colnames(x$inputs$X[,include,drop=F])
if(is.null(names)) names = paste0("Variable", 1:ncol(x$inputs$X[,include,drop=F]))
if(length(names) != length(include)) stop("Number of provided variable names does not match number of included variables.")
#create some global variables such that r cmd check is happy
variable = iter = value = NULL
c.point = apply(x$posterior$beta[,include,,drop=F], c(2,3), mean)
c.upper = apply(x$posterior$beta[,include,,drop=F], c(2,3), quantile, q[2])
c.lower = apply(x$posterior$beta[,include,,drop=F], c(2,3), quantile, q[1])
if(nrow(c.point) == 1){c.upper = matrix(c.upper,nrow=1); c.lower = matrix(c.lower,nrow=1)}
if(is.null(groups)) groups = paste(x$posterior$groups[-ncol(c.point)])
if(length(groups) != ncol(c.point)-1) stop("Wrong number of group names supplied. Need K-1 names where K is the number of choices.")
#kick baseline
c.upper = c.upper[,-ncol(c.upper),drop=F]
c.lower = c.lower[,-ncol(c.lower),drop=F]
c.point = c.point[,-ncol(c.point),drop=F]
#add variable names
c.upper = data.frame(c.upper, names)
c.lower = data.frame(c.lower, names)
c.point = data.frame(c.point, names)
#add group names
colnames(c.upper) = colnames(c.lower) = colnames(c.point) = c(groups, "names")
#add measurement
c.upper$measure = "c.upper"
c.lower$measure = "c.lower"
c.point$measure = "c.point"
plot.df = rbind(c.upper,c.lower,c.point)
plot.df = reshape2::melt(plot.df, id.vars = c("names","measure"))
plot.df = reshape2::dcast(plot.df, "names + variable ~ measure")
#sorting (bit more complicated in MNL, I use average point estimate over all groups)
if(length(names)>1){
if(sort){
average = aggregate(plot.df$c.point, by=list(plot.df$names), FUN=mean)
lvls = unique(plot.df$names)
plot.df$names = factor(plot.df$names, levels = lvls[order(average$x)])
}
}
if(!sort){
# plot in order of appearance in X
plot.df$names = factor(plot.df$names, levels = rev(names))
}
#axis labeling, take care b/c of coord_flip
if(is.null(ylab)) ylab = ""
if(is.null(xlab)) xlab = "Posterior Estimate"
final = ggplot(plot.df, aes(x=names, y=c.point, shape = variable, group = variable)) +
geom_errorbar(aes(x=names, ymin=c.lower, ymax=c.upper, group = variable),
position = position_dodge(width=.5),
width=0, col="grey60") +
geom_point(position = position_dodge(width=.5)) +
theme_bw() +
geom_hline(yintercept = 0,lty=2) +
coord_flip() +
xlab(ylab) +
ylab(xlab) +
theme(legend.position = "bottom",
legend.title = element_blank())
return(final)
}
#' @name plot.UPG.Binomial
#' @title Coefficient plots for UPG.Binomial objects
#'
#' @description \code{plot} generates plots from \code{UPG.Binomial} objects using \code{ggplot2}. Coefficient plots show point estimates for all coefficients as well as their credible intervals.
#'
#' @param x an object of class \code{UPG.Binomial}.
#' @param sort a logical variable indicating whether the plotted coefficients should be sorted according to effect sizes. Default is FALSE.
#' @param names a character vector indicating names for the variables used in the plots.
#' @param xlab a character vector of length 1 indicating a title for the x-axis.
#' @param ylab a character vector of length 1 indicating a title for the y-axis.
#' @param q a numerical vector of length two providing the posterior quantiles to be extracted. Default are 0.025 and 0.975 quantiles.
#' @param include can be used to plot only a subset of variables. Specify the columns of X that should be kept in the plot. See examples for further information.
#' @param ... other plot parameters.
#'
#' @return Returns a ggplot2 object.
#'
#' @seealso
#' \code{\link{summary.UPG.Binomial}} to summarize a \code{UPG.Binomial} object and create tables.
#' \code{\link{predict.UPG.Binomial}} to predict probabilities using a \code{UPG.Binomial} object.
#' \code{\link{coef.UPG.Binomial}} to extract coefficients from a \code{UPG.Binomial} object.
#'
#' @author Gregor Zens
#'
#' @examples
#' \donttest{
#' # estimate a binomial logit model using example data
#' library(UPG)
#' data(titanic)
#' y = titanic[,1]
#' Ni = titanic[,2]
#' X = titanic[,-c(1,2)]
#' results.binomial = UPG(y = y, X = X, Ni = Ni, model = "binomial")
#'
#' # plot the results and sort coefficients by effect size
#' plot(results.binomial, sort = TRUE)
#'
#' # plot only variables 1 and 3 with custom names, credible intervals and axis labels
#' plot(results.binomial,
#' include = c(1,3),
#' names = c("Custom 1", "Custom 2"),
#' q = c(0.1, 0.9),
#' xlab = c("Custom X"),
#' ylab = c("Custom Y"))
#'}
#' @method plot UPG.Binomial
#'
#'@export
plot.UPG.Binomial = function(x = NULL,
...,
sort = FALSE, # sort coefficients by average effect size
names = NULL, # provide names for variables, alternatively
xlab = NULL, # provide x axis label
ylab = NULL, # provide y axis label
q = c(0.025, 0.975), # credible intervals
include = NULL # which variables to include? default:all (numeric vector)
){
if(is.null(include)) include = 1:ncol(x$inputs$X)
if(is.null(names)) names = colnames(x$inputs$X[,include,drop=F])
if(is.null(names)) names = paste0("Variable", 1:ncol(x$inputs$X[,include,drop=F]))
if(length(names) != length(include)) stop("Number of provided variable names does not match number of included variables.")
#create some global variables such that r cmd check is happy
variable = iter = value = NULL
c.point = apply(x$posterior$beta[,include,drop=F], 2, mean)
c.upper = apply(x$posterior$beta[,include,drop=F], 2, quantile, q[2])
c.lower = apply(x$posterior$beta[,include,drop=F], 2, quantile, q[1])
plot.df = data.frame(c.point,c.upper,c.lower,names)
if(nrow(plot.df) > 1 & sort){
plot.df$names = factor(plot.df$names, levels = plot.df$names[order(plot.df$c.point)])
}
if(!sort){
# plot in order of appearance in X
plot.df$names = factor(plot.df$names, levels = rev(plot.df$names))
}
#axis labeling, be careful b/c of coord_flip
if(is.null(ylab)) ylab = ""
if(is.null(xlab)) xlab = "Posterior Estimate"
final = ggplot(plot.df, aes(x=names, y=c.point)) +
geom_errorbar(ymin=c.lower, ymax=c.upper, width=0, col="grey60") +
geom_point() +
geom_point(data = plot.df, aes(x=names, y=c.upper), size=-1) + #for correct scaling
geom_point(data = plot.df, aes(x=names, y=c.lower), size=-1) + #for correct scaling
theme_bw() +
geom_hline(yintercept = 0,lty=2) +
coord_flip() +
xlab(ylab) +
ylab(xlab)
return(final)
}
Try the UPG package in your browser
Any scripts or data that you put into this service are public.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.