A0339
Title: Efficient and sequential estimation of high-order dynamic spatial panels with time-varying strongly dominant units
Authors: Chen Yahui - Xiamen University (China) [presenting]
Han Xiaoyi - Xiamen University (China)
Zhang Jiajun - Shanghai Jiao Tong University (China)
Jin Fei - Fudan University (China)
Abstract: The estimation of a high-order spatial dynamic panel data model is considered with time-varying strongly dominant units and heteroskedasticity. The dominant units vary over time, and the numbers of dominant units are finite or infinite. To accommodate the model specification, a central limit theorem (CLT) is developed where the column sum magnitude in the quadratic form can be equal to one and the existence of heteroskedasticity is allowed. The generalized method of moments estimator (GMME) and root estimator (RE) is proposed, as well as the consistency and asymptotic normality of these estimators when both n and T are large. The advantage of RE is that it has a closed-form solution and is asymptotically as efficient as the best GMME. Monte Carlo simulations demonstrate that the estimators have satisfactory finite sample performances. Finally, an empirical application is presented to illustrate the usefulness of the model on the peer effects of firm finance decisions across Chinese listed firms.
