coeftestIV: Coefficient test of output of ivreg.EX.

Description

For a given set of endogenous variables in an instrumental variables model coeftestIV provides you the coefficients matrix with estimates, std errs, t-ratios and p-values for both homoscedastic and heteroscedastic errors. Also, clustered standard error can be calculated.

Usage

coeftestIV(ivregfit, lmfit, vars, cls = NULL, vcov)

Arguments

ivregfit

An object of class 'ivregEX', an output of ivreg.EX or ivreg.fit.Ex function.

lmfit

An object of class 'lm', an output of lmfit or lm function.

vars

A vector of endogenous variables of the instrumental variables model for which the standard error calculation and significance test to be performed.

cls

A numeric vector specifying the stratification strata structure of the model. Default setting of NULL assumes no stratification.

vcov

A specification of the covariance matrix of the estimated coefficients. This can be specified as a matrix or as a function yielding a matrix when applied to x. Only used for the computation under heteroscedastic errors. Default is vcovHC.

Value

An object of class 'coeftest'. A matrix of coefficient estimates, standard errors, t-ratios and p-values.

Details

For details on vcov see the documentation of coeftest in "lmtest" package.

Examples

Run this code

# NOT RUN {
data("CigarettesSW", package = "AER")
CigarettesSW$rprice <- with(CigarettesSW, price/cpi)
CigarettesSW$rincome <- with(CigarettesSW, income/population/cpi)
CigarettesSW$tdiff <- with(CigarettesSW, (taxs - tax)/cpi)

fm2 <- ivreg.EX(log(packs) ~ log(rprice) + log(rincome) | tdiff + log(rincome),
			data = CigarettesSW, subset = year == "1995")
lmfit <- lm(log(packs) ~ log(rprice) + log(rincome), data = CigarettesSW, 
			subset = year == "1995")

res = coeftestIV(fm2, lmfit, 'log(rprice)')

##############
x = CigarettesSW[CigarettesSW[,'year'] == "1995", c('rprice', 'rincome')]
y = CigarettesSW[CigarettesSW[,'year'] == "1995", c('packs')]
z = CigarettesSW[CigarettesSW[,'year'] == "1995", c('tdiff', 'rincome')]

fm3 <- ivreg.fit.EX(as.matrix(x), y, z)
lmfit <- lm(packs ~ rprice + rincome, data = CigarettesSW, subset = year == "1995")
res = coeftestIV(fm3, lmfit, 'rprice')
res

# }