dot-fcs2SetDefaultPriors: Set Default Priors
In aquaMetrics/fcs2: Fisheries Classification Scheme 2 For SNIFFER

Description Usage Arguments Details Value Note See Also

Sets default prior parameters for variables missing in fit$prior.parameters.

1	.fcs2SetDefaultPriors(fit)

fit

an object of class "fcs2Fit", usually returned by fcs2FitModel.

The default priors are as follows:

The shape parameter r follows a log-Normal distribution with mean=1 and sd=10.
The catch probability q follows a Uniform distribution between 0 and 1.
The constant regression terms beta.const and gamma.const follow a Normal distribution with mean=0 and sd=5.
All other linear terms beta.* or gamma.* follow a Normal distribution with mean=0 and sd=1.
The hyperparameters for random walk terms rw1 or rw2 are specified for the scale parameters sigma.* or nu.* which follow the hyperprior distribution with mean=0.5 and sd=3.
The hyperparameters for spatial terms are specified for the precision parameters tau.* or phi.* and follow a Gamma distribution with shape=1 and rate=0.001.

The priors should always be checked (with plot.fcs2Fit or fcs2Priors) before fitting the full FCS2 model with BUGS and adjusted if necessary through the prior.parameters argument of fcs2FitModel.

the list fit$prior.parameters modified to contain parameters for every variable in the model.

The user should not call this function directly but it is called by fcs2FitModel when creating a complete "fcs2Fit" object.

fcs2FitModel

aquaMetrics/fcs2 documentation built on Aug. 21, 2021, 12:55 p.m.