Abstract
Most Stata commands allow cluster(varname) as an option, popularizing the use of standard errors that are robust to one-way clustering. But
when it comes to adjusting standard errors for multi-way clustering, there is no
solution that is as widely applicable. While several user-written packages support multi-way clustering, each package is compatible with only a subset of models that Stata’s ever-expanding library of commands allows the researcher to estimate. We introduce a command vcemway that provides a one-stop solution for multi-way clustering. vcemway works with any estimation command that allows cluster(varname) as an option, and adjusts standard errors, individual significance statistics and confidence intervals in output tables for multi-way clustering in specified dimensions. The covariance matrix used in making this adjustment is stored in e(V), meaning that any subsequent call to postestimation commands that use e(V) as input (e.g. test and margins) will also produce results that are robust to multi-way clustering.
when it comes to adjusting standard errors for multi-way clustering, there is no
solution that is as widely applicable. While several user-written packages support multi-way clustering, each package is compatible with only a subset of models that Stata’s ever-expanding library of commands allows the researcher to estimate. We introduce a command vcemway that provides a one-stop solution for multi-way clustering. vcemway works with any estimation command that allows cluster(varname) as an option, and adjusts standard errors, individual significance statistics and confidence intervals in output tables for multi-way clustering in specified dimensions. The covariance matrix used in making this adjustment is stored in e(V), meaning that any subsequent call to postestimation commands that use e(V) as input (e.g. test and margins) will also produce results that are robust to multi-way clustering.
Original language | English |
---|---|
Pages (from-to) | 900-912 |
Journal | Stata Journal |
Volume | 19 |
Issue number | 4 |
Early online date | 18 Dec 2019 |
DOIs | |
Publication status | Published - Dec 2019 |