Robust Covariance Matrix Estimators

Model-robust standard error estimators for cross-sectional, time series, clustered, panel, and longitudinal data. Modular object-oriented implementation with support for many model objects, including: lm, glm, survreg, coxph, mlogit, polr, hurdle, zeroinfl, and beyond.

Sandwich covariances for general parametric models:

Object-oriented implementation in R:

library("sandwich")
library("lmtest")
data("PetersenCL", package = "sandwich")
m <- lm(y ~ x, data = PetersenCL)
coeftest(m, vcov = sandwich)

## 
## t test of coefficients:
## 
##             Estimate Std. Error t value Pr(>|t|)    
## (Intercept)   0.0297     0.0284    1.05      0.3    
## x             1.0348     0.0284   36.45   <2e-16 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

coeftest(m, vcov = vcovCL, cluster = ~ firm)

## 
## t test of coefficients:
## 
##             Estimate Std. Error t value Pr(>|t|)    
## (Intercept)   0.0297     0.0670    0.44     0.66    
## x             1.0348     0.0506   20.45   <2e-16 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1