Cluster-robust SEs for regression and bootstrap
Posted: Tue Jan 07, 2025 1:53 pm
Dear PLS-team
A couple years ago, as a smartpls 3 user, I suggested implementing cluster-robust standard errors into SmartPLS (bootstrap). The response was positive, but as I again have become a SmartPLS user, I still can't find that possibility.
Implementing it in the regression toolkit along with other robust standard error adjustments would be natural BUT also in the SEM (PLS and CB) toolkits. Most other (CB) SEM software has that option.
If you have paneldata or data where observations are clustered (in schools, firms, countries, etc) it is essential to be able to implement cluster-robust standard erors rather than only control for fixed effects (dummies).
I am no expert on the coding but this article implies that it is possible to implement clustering with bootstrapping.
Huang, F. L. (2018). Using cluster bootstrapping to analyze nested data with a few clusters. Educational and psychological measurement, 78(2), 297-318.
Regards
Tobias
A couple years ago, as a smartpls 3 user, I suggested implementing cluster-robust standard errors into SmartPLS (bootstrap). The response was positive, but as I again have become a SmartPLS user, I still can't find that possibility.
Implementing it in the regression toolkit along with other robust standard error adjustments would be natural BUT also in the SEM (PLS and CB) toolkits. Most other (CB) SEM software has that option.
If you have paneldata or data where observations are clustered (in schools, firms, countries, etc) it is essential to be able to implement cluster-robust standard erors rather than only control for fixed effects (dummies).
I am no expert on the coding but this article implies that it is possible to implement clustering with bootstrapping.
Huang, F. L. (2018). Using cluster bootstrapping to analyze nested data with a few clusters. Educational and psychological measurement, 78(2), 297-318.
Regards
Tobias