Create a parameter object for the function createSegmentedModel

Source: R/Segmented.R

Create a parameter object for the function createSegmentedModel

createSegmentedArgs(
  modelType,
  seg.Z = ~eventDate,
  psi = NA,
  npsi = NA,
  fixed.psi = NULL,
  control = segmented::seg.control(),
  model = TRUE
)

Arguments

modelType

One of: linear or poisson

seg.Z

the segmented variables(s), i.e. the continuous covariate(s) understood to have a piecewise-linear relationship with response. It is a formula with no response variable, such as seg.Z=~x or seg.Z=~x1+x2.

psi

Starting values for the breakpoints to be estimated. If there is a single segmented variable specified in seg.Z, psi is a numeric vector

npsi

A named vector or list meaning the number (and not locations) of breakpoints to be estimated.

fixed.psi

An optional named list meaning the breakpoints to be kept fixed during the estimation procedure. The names should be a subset of (or even the same) variables specified in seg.Z. If there is a single variable in seg.Z, a simple numeric vector can be specified.

control

a list of parameters for controlling the fitting process. See the documentation for [segmented::seg.control] for details

model

logical value indicating if the model.frame should be returned

Details

Create an object defining the parameter values for use with the createSegmentedModel.

See also

[segmented::segmented()] which this function wraps its arguments.