jabba_hindcast: jabba hindcasting function

In zanbi/jabbamodel-JABBA: Just Another Bayesian Biomass Assessment

Description

Wrapper to coduct histcasts for retrospective analysis and cross-validation

Usage

jabba_hindcast(
  jbinput,
  ni = 30000,
  nt = 5,
  nb = 5000,
  nc = 2,
  init.values = FALSE,
  init.K = NULL,
  init.r = NULL,
  init.q = NULL,
  peels = 0:5,
  save.jabba = FALSE,
  output.dir = getwd(),
  save.hc = FALSE,
  plotall = FALSE,
  speedup = TRUE
)

Arguments

jbinput	List of input variables as output by build_jabba() MCMC settings
ni	number of iterations
nt	thinning interval of saved iterations
nb	burn-in
nc	number of mcmc chains Initial values
init.values	= FALSE,
init.K	= NULL,
init.r	= NULL,
init.q	= NULL,# vector
peels	sequence of retro spective peels default 0:5
save.jabba	Save individual JABBA run outputs
output.dir	path to save plot. default is getwd()
save.hc	Save hindcast list output as .rdata
plotall	if TRUE makes jabba_plots() for each run
speedup	Reduces MCMC after setting runs 2+ inits to first "full" reference run

Value

hc containing estimates of key joint results from all hindcast run

zanbi/jabbamodel-JABBA documentation built on June 27, 2020, 12:23 a.m.