Quick Newton-Raphson with sensible defaults

Convenience wrapper for mle_newton_raphson with simplified interface. Always uses line search for stability (recommended for Newton-Raphson).

Usage

mle_nr(loglike, score, fisher, theta0, inverted = FALSE, ...)

Arguments

loglike

Log-likelihood function

score

Score function

fisher

Fisher information or covariance function

theta0

Initial parameters

inverted

Is fisher the covariance matrix? (default: FALSE)

...

Additional config parameters passed to mle_config_linesearch

Value

mle_newton_raphson object

Examples

if (FALSE) { # \dontrun{
# Quick usage with Fisher information matrix
result <- mle_nr(loglike, score, fisher, theta0 = c(0, 1))

# With covariance matrix (inverted FIM)
result <- mle_nr(
  loglike, score, covariance,
  theta0 = c(0, 1),
  inverted = TRUE
)

# Override config
result <- mle_nr(
  loglike, score, fisher,
  theta0 = c(0, 1),
  max_iter = 50,
  max_step = 0.5
)
} # }