Iteratively reweighted least squares function – newton methods

Iteratively reweighted least squares function – newton methods

Usage

irls_iter_nt(
  y_vec,
  xdumm,
  theta_estimated,
  tolerance = .Machine$double.eps^0.5,
  stop_criteria = 1e-05,
  q_vec
)

Arguments

y_vec

Observed counts

xdumm

Design matrix

theta_estimated

Estimated coefficient values

tolerance

Numeric tolerance for matrix to be considered not revertable, default = .Machine$double.eps^0.5

stop_criteria

Resolution to stop, default = 1e-6

q_vec

Capturing probability vector

Value

Converged coefficient estimate