avar: Asymptotic variance of Estimator

In bchivers-stanford/xtreg2way: Feasible Estimation of the Two-Way Fixed Effect Model

Asymptotic variance of Estimator

Description

avar calculated the asymptotic variance of the regression estimation

Usage

avar(X, e, group = NULL, J = NULL)

Arguments

X	A matrix or vector of independent variable(s)
e	The residuals from the regression
group	(optional) The cluster identifier (hhid from projdummies)
J	(optional) This is assumed to be X'X, and can be input if pre-calculated

Value

A matrix of the covariates

Examples

hhid <- c("a","b","c","a","b","c" ,"a","b","c" ,"a","b","c" ,"a","b","c")
tid <- c("1","1" ,"1" ,"2","2" ,"3","3","3" ,"4","4","5" ,"5","6","6" ,"6")
w <- rep(1, 15)
x1 <- rnorm(15, mean=50, sd=10)   
x2 <- rnorm(15, mean=50, sd=10)
y <- x1 + rnorm(15, mean=50, sd=10)

struc <- projdummies(hhid, tid, w)
projvar_list <- projvar(x1, struc)
x1p <- projvar_list$var
projvar_list <- projvar(x2, struc)
x2p <- projvar_list$var
projvar_list <- projvar(y, struc)
yp <- projvar_list$var

reg <- regress1(yp, data.frame(x1p,x2p))
matCov <- avar(data.frame(x1p, x2p), reg$res, struc$hhid, reg$XX)

bchivers-stanford/xtreg2way documentation built on Jan. 5, 2023, 7:58 p.m.