numDeltaMethod: Utility method to calculate variance using the numerical...

In mclements/numdelta: Numerical delta method

Usage

numDeltaMethod(object, fun, ...)

Arguments

object
fun
...

Examples

##---- Should be DIRECTLY executable !! ----
##-- ==>  Define data, use random,
##--	or do  help(data=index)  for the standard data sets.

## The function is currently defined as
function (object, fun, ...) 
{
    coef <- coef(object)
    Sigma <- vcov(object)
    fit <- fun(coef, ...)
    gd <- grad(fun, coef, ...)
    se.fit <- as.vector(sqrt(diag(t(gd) %*% Sigma %*% gd)))
    names(se.fit) <- names(fit)
    if (all(se.fit == 0)) 
        warning("Zero variance estimated. Do you need to pass a newdata statement?")
    structure(list(fit = fit, se.fit = se.fit), class = "predictnl")
  }

mclements/numdelta documentation built on May 22, 2019, 3:10 p.m.