Skip to contents

Options for the finite-difference covariance in setCov()

Source: R/covControl.R
rsControl.Rd

Used by setCov(fit, "r,s"), "r" and "s". Each option left NULL keeps the value the fit was estimated with, except covFallback.

Usage

rsControl(
  hessEps = NULL,
  gillKcov = NULL,
  gillStepCov = NULL,
  gillFtolCov = NULL,
  covGillF = NULL,
  covSmall = NULL,
  rmatNorm = NULL,
  smatNorm = NULL,
  covFallback = NULL,
  covPrecursor,
  covShortcut = NULL
)

Arguments

hessEps

is a double value representing the epsilon for the Hessian calculation. This is used for the R matrix calculation.

gillKcov

Max steps to determine the optimal forward/central difference step size per parameter (Gill 1983) during the covariance step. `0` = no optimal step size determined.

gillStepCov

When looking for the optimal forward difference step size, this is This is the step size to increase the initial estimate by. So each iteration during the covariance step is equal to the new step size = (prior step size)*gillStepCov

gillFtolCov

The gillFtol is the gradient error tolerance that is acceptable before issuing a warning/error about the gradient estimates during the covariance step.

covGillF

Use the Gill calculated optimal Forward difference step size for the instead of the central difference step size during the central difference gradient calculation of the theta-only stage (see covFull).

covSmall

Small number used to compare covariance estimates (sandwich vs R/S matrix) before rejecting one as too small to be the final covariance estimate.

rmatNorm

A parameter to normalize gradient step size by the parameter value during the calculation of the R matrix of the theta-only stage (see covFull)

smatNorm

A parameter to normalize gradient step size by the parameter value during the calculation of the S matrix of the theta-only stage (see covFull)

covFallback

what the requested method may fall back to, as in foceiControl(covFallback=). NULL (default) is none: setCov() installs the method asked for or stops with an error. With a list, a covariance from a method it lists for the request is installed, with a warning that names it.

covPrecursor

what the covariance may start from, as in foceiControl(covPrecursor=). Left out, the fit's value is kept; NULL uses none.

covShortcut

as in foceiControl(covShortcut=); NULL (default) keeps the fit's value.

Value

rsControl object

See also

setCov()

Author

Matt Fidler

Examples

rsControl(hessEps = 1e-4)
#> $hessEps
#> [1] 1e-04
#> 
#> attr(,"class")
#> [1] "rsControl"