hyperpar: Improved estimates for the hyperparameters
In andrewzm/INLA: Functions which allow to perform full Bayesian analysis of latent Gaussian models using Integrated Nested Laplace Approximaxion

Description Usage Arguments Value Note Author(s) References See Also

Improve the estimates of the posterior marginals for the hyperparameters of the model using the grid integration strategy.

 inla.hyperpar(
         result,
         skip.configurations = TRUE,
         verbose = FALSE,
         dz = 0.75,
         diff.logdens = 7,
         h = NULL,
         restart = FALSE,
         quantiles = NULL, 
         keep = FALSE)

`result`	An object of class `inla`, ie a result of a call to `inla()`
`skip.configurations`	A boolean variable; skip configurations if the values at the main axis are to small. (Default TRUE)
`verbose`	Boolean indicating wheather the inla program should run in a verbose mode.
`dz`	Step length in the standardized scale used in the construction of the grid, default 0.75.
`diff.logdens`	The difference of the log.density for the hyperpameters to stop numerical integration using int.strategy='grid'. Default 7
`h`	The step-length for the gradient calculations for the hyperparameters. Default 0.01.
`restart`	A boolean defining wheather the optimizer should start again to ind the mode or if it should use the mode contained in the `object`
`quantiles`	A vector of quantiles, to compute for each posterior marginal.
`keep`	A boolean variable indicating the working files (ini file, data files and results files) should be kept

The object returned is the same as object but the estimates of the hyperparameters are replaced by improved estimates.

This function might take a long time or if the number of hyperparameters in the model is large. If it complains and says I cannot get enough memory, try to increase the value of the argument dz or decrease diff.logdens.

Havard Rue hrue@math.ntnu.no

See the references in inla

inla

andrewzm/INLA documentation built on May 10, 2019, 11:12 a.m.