mscjs_tmb function

Fitting function for Multistate CJS models with TMB

A function for computing MLEs for a Multi-state Cormack-Jolly-Seber open population capture-recapture model for processed dataframe x with user specified formulas in parameters that create list of design matrices dml. This function can be called directly but is most easily called from crm that sets up needed arguments.

mscjs_tmb(
  x,
  ddl,
  fullddl,
  dml,
  model_data = NULL,
  parameters,
  accumulate = TRUE,
  initial = NULL,
  method,
  hessian = FALSE,
  debug = FALSE,
  chunk_size = 1e+07,
  refit,
  itnmax = NULL,
  control = NULL,
  scale,
  re = FALSE,
  compile = FALSE,
  extra.args = "",
  clean = TRUE,
  getreals = FALSE,
  useHess = FALSE,
  savef = TRUE,
  ...
)

Arguments

• x: processed dataframe created by process.data
• ddl: list of simplified dataframes for design data; created by call to make.design.data
• fullddl: list of complete dataframes for design data; created by call to make.design.data
• dml: list of design matrices created by create.dm from formula and design data
• model_data: a list of all the relevant data for fitting the model including imat, S.dm,p.dm,Psi.dm,S.fixed,p.fixed,Psi.fixed and time.intervals. It is used to save values and avoid accumulation again if the model was re-rerun with an additional call to cjs when using autoscale or re-starting with initial values. It is stored with returned model object.
• parameters: equivalent to model.parameters in crm
• accumulate: if TRUE will accumulate capture histories with common value and with a common design matrix all parameters to speed up execution
• initial: list of initial values for parameters if desired; if each is a named vector from previous run it will match to columns with same name
• method: method to use for optimization; see optim
• hessian: if TRUE will compute and return the hessian
• debug: if TRUE will print out information for each iteration
• chunk_size: specifies amount of memory to use in accumulating capture histories; amount used is 8*chunk_size/1e6 MB (default 80MB)
• refit: non-zero entry to refit
• itnmax: maximum number of iterations
• control: control string for optimization functions
• scale: vector of scale values for parameters
• re: if TRUE creates random effect model admbcjsre.tpl and runs admb optimizer
• compile: if TRUE forces re-compilation of tpl file
• extra.args: optional character string that is passed to tmb
• clean: if TRUE, deletes the dll and recompiles
• getreals: if TRUE, compute real values and std errors for TMB models; may want to set as FALSE until model selection is complete
• useHess: if TRUE, the TMB hessian function is used for optimization; using hessian is typically slower with many parameters but can result in a better solution
• savef: if TRUE, save the makeAdFun result from TMB to report real values and matrices
• ...: not currently used

Returns

The resulting value of the function is a list with the class of crm,cjs such that the generic functions print and coef can be used. - beta: named vector of parameter estimates - lnl: -2log likelihood - AIC: lnl + 2 number of parameters

• convergence: result from optim; if 0 optim thinks it converged - count: optim results of number of function evaluations - reals: dataframe of data and real S and p estimates for each animal-occasion excluding those that occurred before release

• vcv: var-cov matrix of betas if hessian=TRUE was set

Details

It is easiest to call mscjs_tmb through the function crm. Details are explained there.

References

Ford, J. H., M. V. Bravington, and J. Robbins. 2012. Incorporating individual variability into mark-recapture models. Methods in Ecology and Evolution 3:1047-1054.

Author(s)

Jeff Laak