Question

To help my understanding of Bayesian updating I have been working with the code from bayesianbiologist. As I have to learn how to create animated plots, I thought it would be a fun exercise to create an animated plot of the updating. This has proved tougher than expected. Taking inspiration from Rob Hyndman’s blog on this issue I tried to create the following:

library(animation)
setwd("~/Dropbox/PriorUpdating") #set working directory

## Simulate Bayesian Binomial updating

sim_bayes<-function(p=0.5,N=10,y_lim=15,prior_a=1,prior_b=1)
{
   print(paste("The prior expectation of p is ",prior_a/(prior_a+prior_b)))
   success<-0
   curve(dbeta(x,prior_a,prior_b),xlim=c(0,1),ylim=c(0,y_lim),xlab='p',ylab='Posterior Density',lty=2)
   legend('topright',legend=c('Prior','Updated Posteriors','Final Posterior'),lty=c(2,1,1),col=c('black','black','red'))

  for(i in 1:N)
     {

        if(runif(1,0,1)<=p) success<-success+1 #this is where we see if there is a "success"

      curve(dbeta(x,success+prior_a,(i-success)+prior_b),add=TRUE) #plot updated
      }
curve(dbeta(x,success+prior_a,(i-success)+prior_b),add=TRUE,col='red',lwd=1.5) #plot final posterior
}

oopt = ani.options(interval = 0)
saveMovie(sim_bayes(p=0.6,N=90,prior_a=1,prior_b=1),interval=0.1,width=580,height=400)
ani.options(oopt)

This however only produced the final plot. So I thought I would try to output all of the PDFs of the plots.

library(animation)
setwd("~/Dropbox/PriorUpdating") #set working directory

## Simulate Bayesian Binomial updating

sim_bayes<-function(p=0.5,N=10,y_lim=15,prior_a=1,prior_b=1)
{
  print(paste("The prior expectation of p is ",prior_a/(prior_a+prior_b)))
  success<-0



  curve(dbeta(x,prior_a,prior_b),xlim=c(0,1),ylim=c(0,y_lim),xlab='p',ylab='Posterior Density',lty=2)
  legend('topright',legend=c('Prior','Updated Posteriors','Final Posterior'),lty=c(2,1,1),col=c('black','black','red'))



  for(i in 1:N)
  {
    pdf(paste("posterior",i,".pdf",sep=""),height=4,width=6.5)

    if(runif(1,0,1)<=p) success<-success+1 #this is where we see if there is a "success"

    curve(dbeta(x,success+prior_a,(i-success)+prior_b),add=TRUE) #plot updated

    #print(paste(success,"successes and ",i-success," failures"))
    dev.off()
  }
  pdf(paste("posterior_final",".pdf",sep=""),height=4,width=6.5)
  curve(dbeta(x,success+prior_a,(i-success)+prior_b),add=TRUE,col='red',lwd=1.5) #plot final posterior
  dev.off()
}

However this gives me the following error

Error in plot.xy(xy.coords(x, y), type = type, ...) : 
  plot.new has not been called yet

I tried inserting plot.new() at certain points but I think this conflicts with the additive nature of the plots.

Has anyone got any idea how I could make one or both of these methods work properly? While this is a bit of a toy example for me, I have some more interesting animations I need to plot where understanding how to animate plots is going to be useful/necessary.

Thanks for any help!

Answer 1

Like MatthewK said, you’ve misplaced the call to pdf. However, this should get you up and running:

sim_bayes <- function(p=0.5, N=10, y_lim=15, prior_a=1, prior_b=1) {
    success <- 0
    for (i in 1:N) {
        pdf(paste("posterior",i,".pdf",sep=""), height=4, width=6.5)

        if (runif(1,0,1) <= p)
            success<-success + 1

        # Start a new plot.
        curve(dbeta(x,prior_a,prior_b), lty=2,
              xlim=c(0,1), ylim=c(0,y_lim), xlab='p', ylab='Posterior Density')
        # Update plot.
        curve(dbeta(x,success+prior_a, (i-success) + prior_b), add=TRUE)

        legend('topright',
               legend=c('Prior','Updated Posteriors','Final Posterior'),
               lty=c(2,1,1), col=c('black','black','red'))

        dev.off()
    }
}

# `x` had no visible binding in your implementation, so I took the following
# from the `dbeta` documentation example.
x <- seq(0, 1, length=21)
sim_bayes()