StochasticSRA: Stochastic SRA construction of operating models

Description Usage Arguments Value Author(s) References Examples

View source: R/StochasticSRA.R

Description

Specify an operating model, using catch composition data and a historical catch series. Returns and operating model with depletion (D), selectivity parameters (L5, LFS) and effort trajectory (Effyears, EffLower, EffUpper) filled.

Usage

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
StochasticSRA(
  OM,
  CAA,
  Chist,
  Ind = NA,
  ML = NA,
  CAL = NA,
  mulen = NA,
  wts = c(1, 1, 0.5, 0.1, 1),
  Jump_fac = 1,
  nits = 4000,
  burnin = 500,
  thin = 10,
  ESS = 300,
  MLsd = 0.1,
  ploty = T,
  nplot = 6,
  SRAdir = NA
)

Arguments

OM

An operating model object with M, growth, stock-recruitment and maturity parameters specified.

CAA

A matrix nyears (rows) by nages (columns) of catch at age (age 1 to maxage in length)

Chist

A vector of historical catch observations (nyears long) going back to unfished conditions

Ind

A vector of historical abundance index observations (assumed proportional to SSB)

ML

A vector of historical mean length (in catch) observations

CAL

A matrix of nyears (row) by n length bins (columns) of catch at length samples

mulen

A vector mean length by length bin, a vector the same as the number of columns of CAL

wts

A vector of relative weights for the likelihood functions of CAA, Chist, Ind, ML and CAL

Jump_fac

A multiplier of the jumping distribution variance to increase acceptance (lower Jump_fac) or decrease acceptance rate (higher Jump_fac)

nits

The number of MCMC iterations

burnin

The number of initial MCMC iterations to discard

thin

The interval over which MCMC samples are extracted for use in graphing / statistics

ESS

Effective sample size - the weighting of the catch at age data

MLsd

The lognormal sd of the mean length observations

ploty

Do you want to see diagnostics plotted?

nplot

how many MCMC samples should be plotted in convergence plots?

SRAdir

A directory where the SRA diagnostics / fit are stored

Value

A list with three positions. Position 1 is the filled OM object, position 2 is the custompars data.frame that may be submitted as an argument to runMSE() and position 3 is the matrix of effort histories [nyears x nsim] vector of objects of classclassy

Author(s)

T. Carruthers (Canadian DFO grant)

References

Walters, C.J., Martell, S.J.D., Korman, J. 2006. A stochastic approach to stock reduction analysis. Can. J. Fish. Aqua. Sci. 63:212-213.

Examples

1
2
3
4
5
6
7
8
9
## Not run: 
setup()
sim<-SRAsim(testOM,patchy=0.8)
CAA<-sim$CAA
Chist<-sim$Chist
testOM<-StochasticSRA(testOM,CAA,Chist,nsim=30,nits=1000)
runMSE(testOM)

## End(Not run)

DLMtool documentation built on March 13, 2020, 2:52 a.m.