Paper ID
1301
Paper Type
full
Description
As a quick econometric solution to potential endogeneity issues in panel data models, the generalized method of moments (GMM) estimator is gaining popularity in IS research. Despite the sensitivity of this estimator to model specifications and estimation strategies, a noticeable number of IS studies employing this method fail to report the detailed model specifications, robustness check results with different specifications and estimation strategies, or test statistics, which render their empirical results less credible. We demonstrate that passing the commonly required tests such as the m2 test and the Sargan-Hansen test does not guarantee the validity of the estimate, because the size and statistical significance of the estimate can depend on the choice of estimation procedure and moment restrictions that pass such required tests. We urge researchers to be explicit about the model specifications and estimation strategies, and to provide robustness checks with different model specifications, along with complete test results.
Recommended Citation
Cheng, Ningning and Bang, Youngsok, "A Comment on GMM Estimation in IS Research" (2019). ICIS 2019 Proceedings. 2.
https://aisel.aisnet.org/icis2019/research_methods/research_methods/2
A Comment on GMM Estimation in IS Research
As a quick econometric solution to potential endogeneity issues in panel data models, the generalized method of moments (GMM) estimator is gaining popularity in IS research. Despite the sensitivity of this estimator to model specifications and estimation strategies, a noticeable number of IS studies employing this method fail to report the detailed model specifications, robustness check results with different specifications and estimation strategies, or test statistics, which render their empirical results less credible. We demonstrate that passing the commonly required tests such as the m2 test and the Sargan-Hansen test does not guarantee the validity of the estimate, because the size and statistical significance of the estimate can depend on the choice of estimation procedure and moment restrictions that pass such required tests. We urge researchers to be explicit about the model specifications and estimation strategies, and to provide robustness checks with different model specifications, along with complete test results.