meat: Clustered meat matrix for a sandwich estimator

In josherrickson/epb: Peters-Belson with Prognostic Heterogeneity

Clustered meat matrix for a sandwich estimator

Description

This meat function is an overloaded function of meat from the sandwich package which enables a clusters argument.

Usage

meat(x, adjust = FALSE, cluster = NULL, ...)

Arguments

x	a fitted model object.
adjust	See sandwich::meat. If a cluster is given, this argument is ignored as the adjustment is always used. The adjustment is m/(m - 1) * (n - 1)/(n - k), where n is the number of observations, k the number of estimated parameters, and m the number of clusters. Note that if there are no clusters (which can be thought of as each observation being its own cluster), m = n and this reduces to n/(n-k), the default in sandwich::meat. See Cameron, A Colin, and Douglas L Miller. 2010. "Robust Inference with Clustered Data." Working Papers, University adjustment used when clustering. of California, Department of Economics.
cluster	A vector identifying cluster membership
...	Additional arguments to sandwich::estfun.

Value

A meat matrix.

See Also

sandwich::meat

josherrickson/epb documentation built on July 6, 2023, 9:12 p.m.