posterior.plot: Posterior distribution plots for summary estimates and...

In giabaio/bmeta: Bayesian Meta-Analysis and Meta-Regression

Posterior distribution plots for summary estimates and between-study standard deviation (measurement of heterogeneity)

Description

Function to create posterior distribution plots for summary estimates and between-study standard deviation based on output from bmeta

Usage

posterior.plot(
  x,
  xlim = NULL,
  xlab = "",
  main = "Posterior distribution Plot",
  scale = "log",
  heterogeneity = FALSE
)

Arguments

x	a bmeta object with results of the model
xlim	horozontal limits of the plot region. If unspecified, the function sets the horizontal plot limits to (-3,3) for binary and count data and (-5,5) for continuous data.
xlab	title for the x-axis
main	title of the plot. If unspecified, the function sets an appropriate title by default.
scale	logical specifying whether summary estimates need to be displayed on log ("log") or natural scale ("exp"). For continuous data, summary estimates are always displayed on natural scale, therefore, users do not need to specify this option.
heterogeneity	logical specifying whether to resent posterior plot for between-study standard deviation (TRUE) to examine heterogeneity of different studies. If unspecified, FALSE by default.

Author(s)

Tao Ding Gianluca Baio

References

Anzures-Cabrera,J & Higgins,J.P.T.(2010) Graphical displays for meta-analysis: An overview with suggestions for practice.Res Synth Methods,1,66-80.

Examples

### Read and format the data (binary)
data = read.csv(url("https://gianluca.statistica.it/software/bmeta/Data-bin.csv"))

### List data for binary outcome 
data.list <- list(y0=data$y0,y1=data$y1,n0=data$n0,n1=data$n1)

### Select random-effects meta-analysis with t-distribution prior for binary
### data
x <- bmeta(data.list, outcome="bin", model="std.dt", type="ran")

### using output from bmeta to produce posterior plot
posterior.plot(x)

### using output from bmeta and specify the horizontal limits 
posterior.plot(x,xlim=c(-2,1))

### using output from bmeta on natural scale and specify more options 
posterior.plot(x,xlim=c(-0.5,2.5),xlab="odds ratio",main="Posterior distribution
of pooled odds ratio", scale="exp")

### examine heterogeneity by producing posterior plot for between-study standard
### deviation
posterior.plot(x,heterogeneity=TRUE,xlim=c(0,3),xlab="between-study standard 
deviation")

giabaio/bmeta documentation built on May 4, 2023, 3:31 p.m.