mkfun: Deriving the log-lik and gradients

In queezzz/qzmle: Maximum likelihood estimation

Description

Deriving the log-lik and gradients

Usage

mkfun(formula, start, links = NULL, parameters = NULL, data)

Arguments

formula	A formula in expression form of "y ~ model"
start	A list of start values for formula parameters
links	Link function for each parameters
parameters	A list of linear submodels
data	A list of parameter in the formula with values in vectors

Examples

set.seed(101)
dd <- data.frame(y = rpois(100, lambda = 1))
fun1 <- mkfun(y ~ dpois(exp(lambda)), start = list(lambda = 0), data = dd)
fun2 <- mkfun(y ~ dnorm(mean = b0 + b1 * latitude^2, sd = 1),start = list(lambda = 0), data = dd)
rfp <- transform(emdbook::ReedfrogPred, nsize = as.numeric(size), random = rnorm(48))
form <- surv ~ dbinom(size = density, prob = exp(log_a) / (1 + exp(log_a) * h * density))
fun3 <- mkfun(form, start = list(h = 1, log_a = 0),
                parameters = list(log_a ~ poly(nsize)), data = rfp)
fun4 <- mkfun(form, start = list(h = 4, log_a = 2),
                parameters = list(log_a ~ poly(random)), data = rfp)

queezzz/qzmle documentation built on Nov. 24, 2021, 6:34 p.m.