marginal_statistics {ArchaeoPhases} | R Documentation |
Calculates summary statistics of the output of the MCMC algorithm for a single parameter. Results are given in calendar years (BC/AD).
marginal_statistics(a_chain, level = 0.95, round_to = 0)
a_chain |
Numeric vector containing the output of the MCMC algorithm for the parameter. |
level |
Probability corresponding to the level of confidence used for the credible interval and the highest posterior density region. |
round_to |
Integer indicating the number of decimal places. |
The (100 * level)
\
using hdr()
function from hdrcde package.
A list with the following components:
The mean of the MCMC chain.
The maximum a posteriori of the MCMC chain.
The standard deviation of the MCMC chain.
A vector with the following elements:
min
= minimum value of the MCMC chain;
q1
= first quantile of the MCMC chain;
median
= median of the MCMC chain;
q2
= second quantile of the MCMC chain; and
max
= maximum value of the MCMC chain.
Confidence level for the credible interval and highest posterior density.
A vector with the following elements:
inf
= lower credible interval of the MCMC chain at level
; and
sup
= upper credible interval of the MCMC chain at level
.
A variable length vector with the lower and upper highest
posterior density regions of the MCMC chain at level
. List
components are named inf_n
and sup_n
for n = 1 to the
number of highest posterior density regions.
Anne Philippe, Anne.Philippe@univ-nantes.fr,
Marie-Anne Vibet, Marie-Anne.Vibet@univ-nantes.fr, and
Thomas S. Dye, tsd@tsdye.online
Hyndman, R. J. (1996) Computing and graphing highest density regions. American Statistician, 50, 120-126.
data(Events); attach(Events)
marginal_statistics(Event.1)
marginal_statistics(Event.2, level = 0.90)
## convenient vector
foo <- marginal_statistics(Event.1)
unlist(foo)