plot.sns: Plotting "sns" Objects

In sns: Stochastic Newton Sampler (SNS)

Plotting "sns" Objects

Description

Method for visualizing the output of sns.run.

Usage

## S3 method for class 'sns'
plot(x, nburnin = max(nrow(x)/2, attr(x, "nnr"))
  , select = if (length(x) <= 10) 1:5 else 1, ...)

Arguments

x	Object of class "sns", typically the output of sns.run.
nburnin	Number of burn-in iterations to discard before generating effective sample size, histograms, and autocorrelation plots.
select	Which plot types must be generated. See below for description.
...	Arguments passed to/from other functions.

Value

plot.sns produces the following types of plots: 1) log-probability trace plot (vertical line, if present, indicates transition from nr to mcmc mode), 2) trace plot of state variables (one per coordinate; vertical line has same meaning as 1), 3) effective sample size by coordinate (horizontal line indicates maximum effective size possible, equal to number of samples after discarding nburnin initial iterations), 4) post-burnin state vector histograms (one per coordinate, vertical line indicates post-burnin average, 5) autocorrelation plots, one per coordinate.

Author(s)

Alireza S. Mahani, Asad Hasan, Marshall Jiang, Mansour T.A. Sharabiani

References

Mahani A.S., Hasan A., Jiang M. & Sharabiani M.T.A. (2016). Stochastic Newton Sampler: The R Package sns. Journal of Statistical Software, Code Snippets, 74(2), 1-33. doi:10.18637/jss.v074.c02

See Also

sns.run

sns documentation built on Nov. 2, 2022, 5:15 p.m.