Plots: ggplot2 for RStan

Description

Visual posterior analysis using ggplot2.

Usage

stan_plot(object, pars, include = TRUE, unconstrain = FALSE,
            ...)
            
  stan_trace(object, pars, include = TRUE, unconstrain = FALSE,
            inc_warmup = FALSE, nrow = NULL, ncol = NULL,
            ...)
  stan_scat(object, pars, include = TRUE, unconstrain = FALSE,
            inc_warmup = FALSE, nrow = NULL, ncol = NULL,
            ...)
  stan_hist(object, pars, include = TRUE, unconstrain = FALSE,
            inc_warmup = FALSE, nrow = NULL, ncol = NULL,
            ...)
  stan_dens(object, pars, include = TRUE, unconstrain = FALSE,
            inc_warmup = FALSE, nrow = NULL, ncol = NULL,
            ..., separate_chains = FALSE)
            
  stan_ac(object, pars, include = TRUE, unconstrain = FALSE,
            inc_warmup = FALSE, nrow = NULL, ncol = NULL,
            ..., separate_chains = FALSE, lags = 25, partial = FALSE)
            
  quietgg(gg)

Arguments

object

A stanfit or stanreg object.

pars

Optional character vector of parameter names. If object is a stanfit object, the default is to show all user-defined parameters or the first 10 (if there are more than 10). If object is a stanreg object, the default

include

Should the parameters given by the pars argument be included (the default) or excluded from the plot?

unconstrain

Should parameters be plotted on the unconstrained space? Defaults to FALSE. Only available if object is a stanfit object.

inc_warmup

Should warmup iterations be included? Defaults to FALSE.

nrow,ncol

Passed to facet_wrap.

...

Additional named arguments to geoms (e.g. for stan_trace the geom is geom_path and we could specify linetype, size, alpha, etc.). For s

separate_chains

For stan_dens, should the density for each chain be plotted? The default is FALSE, which means that for each parameter the draws from all chains are combined. For stan_ac, if separate_chains=FALSE

lags

For stan_ac, the maximum number of lags to show.

partial

For stan_ac, should partial autocorrelations be plotted instead? Defaults to FALSE.

A ggplot object or an expression that creates one.

Value

A ggplot object that can be further customized using the ggplot2 package.

Details

For stan_plot, there are additional arguments (listed below with their default values) that can be specified in ....

[object Object],[object Object],[object Object],[object Object],[object Object],[object Object]

Examples

Run this code

example("read_stan_csv")
stan_plot(fit)
stan_trace(fit)

library(gridExtra)
fit <- stan_demo("eight_schools")

stan_plot(fit)
stan_plot(fit, show_density = TRUE, fill_color = "maroon")


# histograms
stan_hist(fit)
# suppress ggplot2 messages about default bindwidth
quietgg(stan_hist(fit))
quietgg(h <- stan_hist(fit, pars = "theta", binwidth = 5)) 

# juxtapose histograms of tau and unconstrained tau 
tau <- stan_hist(fit, pars = "tau")
tau_unc <- stan_hist(fit, pars = "tau", unconstrain = TRUE) + 
            xlab("tau unconstrained")
grid.arrange(tau, tau_unc)

# kernel density estimates
stan_dens(fit)
(dens <- stan_dens(fit, fill = "skyblue", ))
dens <- dens + ggtitle("Kernel Density Estimates\n") + xlab("")
dens

(dens_sep <- stan_dens(fit, separate_chains = TRUE, alpha = 0.3))
dens_sep + scale_fill_manual(values = c("red", "blue", "green", "black"))
(dens_sep_stack <- stan_dens(fit, pars = "theta", alpha = 0.5,
                             separate_chains = TRUE, position = "stack"))

# traceplot
trace <- stan_trace(fit)
window <- coord_cartesian(xlim = c(1500,1600))
quietgg(trace + window)

trace +
  scale_color_manual(values = c("red", "blue", "green", "black"))

trace +
  scale_color_brewer(type = "div") +
  theme(legend.position = "none")

facet_style <- theme(strip.background = element_rect(fill = "white"),
                     strip.text = element_text(size = 13, color = "black"))
(trace <- trace + facet_style)

# scatterplot
(mu_vs_tau <- stan_scat(fit, pars = c("mu", "tau"), color = "blue", size = 4))
mu_vs_tau + 
  coord_flip() + 
  theme(panel.background = element_rect(fill = "black"))

Run the code above in your browser using DataLab

State of Data and AI Literacy Report 2025