Plot.MixReparametrized: plot of the MCMC output produced by K.MixReparametrized

Description

This is a generic function for a graphical rendering of the MCMC samples produced by K.MixReparametrized function. The function draws boxplots for unimodal variables and for multimodal arguments after clustering them by applying a k-means algorithm. It also plots line charts for other variables.

Usage

Plot.MixReparametrized(xobs, estimate)

Arguments

xobs

vector of the observations

estimate

output of the K. MixReparametrized function

Value

boxplot: three boxplots for the radial coordinates, the mean and the standard deviation of the mixture distribution, $k$ boxplots for each of the mixture model weights, component means and component standard deviations.
histogram: an histogram of the observations against an overlaid curve of the density estimate, obtained by averaging over all mixtures corresponding to the MCMC draws,
line chart: line charts that report the evolution of the proposal scales and of the acceptance rates over the number of batch of 50 iterations.

Details

Boxplots are produced using the boxplot.default method.

References

Kamary, K., Lee, J.Y., and Robert, C.P. (2015) Non-informative reparameterisation of location-scale mixtures. arXiv.

Examples

Run this code

data(faithful)
xobs=faithful[50:100,1]
#estimate=K.MixReparametrized(xobs, k=2, alpha0=.5, alpha=.5, Nsim=1e4)
#plo=Plot.MixReparametrized(xobs, estimate)

Run the code above in your browser using DataLab