output.glm.control {geoRglm} | R Documentation |
This auxiliary function defines output options for
pois.krige.bayes
and binom.krige.bayes
.
output.glm.control(sim.posterior, sim.predict, keep.mcmc.sim, quantile, threshold, inference, messages)
sim.posterior |
logical. Indicates
whether or not the MCMC-sample from the posterior distribution of
the parameters should be returned. Default is sim.posterior = TRUE . |
sim.predict |
logical. Defines whether simulations are
drawn from the predictive distribution. Only valid if prediction
locations are provided in the argument locations . Default is sim.predict = FALSE . |
keep.mcmc.sim |
logical. Indicates
whether or not the MCMC-sample from the posterior distribution of
the signal should be returned. Here the signal
is the inverse link function of the random effect
g^{-1}(S), which equals
exp(S) for the Poison model with log-link,
(lambda S + 1)^{lambda} for the Poison model with Box-Cox-link
and exp(S)/(1+exp(S)) for the binomial model with logit-link. Default is
keep.mcmc.sim = TRUE . |
quantile |
indicates whether quantiles of the
simulations from the
predictive distributions are computed and returned.
If a vector with numbers in the interval
[0,1] is provided, the output includes the object
quantile , which contains values of corresponding
estimated quantiles. For
example, if quantile = c(0.25,0.50,0.75) the function
returns the quartiles of the distributions at each of the
prediction locations.
The default is quantile = TRUE where the values c(0.025, 0.5, 0.975) are used.
If quantile = FALSE no quantiles are computed (and hence
neither median nor uncertainty is included in the output).
Only used if prediction locations are provided in the argument
locations . |
threshold |
one or more values with threshold values can be given to this argument.
If so, an object called
probability is included in the output. This object contains, for each
prediction location, the probability that the variable is less
than or equal to the threshold value given in the argument. Default is that
nothing is computed. |
inference |
logical. Indicates whether or not inference
(summary of the parameters, and prediction) is performed or
not. Default is inference=TRUE .
Setting inference=FALSE is useful in an initial stage, when
tuning the MCMC algorithm (choosing S.scale and phi.scale appropriate).
|
messages |
logical. Indicates whether or not status messages are printed on the screen (or other output device) while the function is running. |
A list with processed arguments to be passed to the main function.
Ole F. Christensen OleF.Christensen@agrsci.dk,
Paulo J. Ribeiro Jr. Paulo.Ribeiro@est.ufpr.br.
pois.krige.bayes
and binom.krige.bayes
.