summary.pogit: Summary for posterior of a `pogit` object

Description

Returns basic information about the model and the priors, MCMC details and (model averaged) posterior means with 95%-HPD intervals for the regression effects and estimated posterior inclusion probabilities.

Usage

"summary"(object, IAT = FALSE, printRes = FALSE, ...)
"print"(x, ...)

Arguments

object

an object of class pogit

IAT

if TRUE, integrated autocorrelation times (IAT) and effective samples sizes (ESS) of the MCMC samples are computed (see details); defaults to FALSE.

printRes

if TRUE, model averaged posterior means for the reporting probabilities and risks are computed for the Pogit model; defaults to FALSE.

...

further arguments passed to or from other methods (not used)

a summary.pogit object produced by summary.pogit()

Value

an object of class summary.pogit

Details

To assess mixing and efficiency of MCMC sampling, the effective sample size (ESS) and the integrated autocorrelation time (IAT) are computed. ESS estimates the equivalent number of independent draws corresponding to the dependent MCMC draws and is defined as ESS = $M$/$\tau$, where $\tau$ is the IAT and $M$ is the number of MCMC iterations after the burn-in phase. IAT is computed as $\tau = 1 + 2 \sum_{k=1}^K \rho(k)$ using the initial monotone sequence estimator (Geyer, 1992) for K and $\rho(k)$ is the empirical autocorrelation at lag $k$.

References

Geyer, C. J. (1992). Practical Markov Chain Monte Carlo. Statistical Science, 7, 473-483.