gradtheta_bes

vector of parameters (all coefficients: kappa and lambda).

theta

parameter representing E(1/W_i|Z_i = z_i, theta).

response vector with 0 &lt; z_i &lt; 1.

matrix containing the covariates for the mean submodel. Each column is a different covariate.

matrix containing the covariates for the precision submodel. Each column is a different covariate.

a string containing the link function for the mean.
The possible link functions for the mean are "logit","probit", "cauchit", "cloglog".

link.mean

a string containing the link function the precision parameter.
The possible link functions for the precision parameter are "identity", "log", "sqrt", "inverse".

link.precision

Function to calculate the gradient of the Q-function, which is required for optimization via <code>optim</code>.
This option is related to the bessel regression.

Functions to fit, via Expectation-Maximization (EM) algorithm, the Bessel and Beta regressions to a data set with a bounded continuous response variable. The Bessel regression is a new and robust approach proposed in the literature. The EM version for the well known Beta regression is another major contribution of this package. See details in the references Barreto-Souza, Mayrink and Simas (2022) <doi:10.1111/anzs.12354> and Barreto-Souza, Mayrink and Simas (2020) <arXiv:2003.05157>.

gradtheta_bes: gradtheta_bes

Description

Usage

Arguments

Value