mr_optim_pred: Predictions and summaries from mr_optim_fit

In daktx2/MSSS: Fits multiresolution spatial models with spatially varying resolution

Description

Predictions and summaries from mr_optim_fit

Usage

mr_optim_pred(locations, results, design_mat = NULL,
  marg_type = "bic_CAP", int_type = "pred", level = 0.95,
  a_beta = 1, b_beta = 1, a_g = 3)

Arguments

locations	1d vector or 2d matrix of locations of the locations you'd like to predict at
results	output from mr_optim_fit
design_mat	if fixed effects are part of the model, design matrix for the locations
marg_type	howt to calculate the marginal, "bic_CAP" is default and strongly recommended but "hyperg_refit" is an option
int_type	"pred" for prediction, "recount" to count resolutions, "noint" for a prediction only
level	confidence level for the interval
a_beta	a parameter for beta prior on model space, 1 is default
b_beta	b parameter for beta prior on model space, 1 is default, but if data is large, use 10^(3*n/10^4)
a_g	if "hyperg_refit" is used

Value

list with either a row matrix of predictions or a row matrix of resolutions active FILLER

daktx2/MSSS documentation built on May 24, 2020, 4:28 a.m.