Accessibility statement

A Unified Approach to Efficient Estimation of Short LinearPanel Regression Models (joint with Takashi Yamagata)

Thursday 13 February 2020, 1.00PM to - 2.00 pm

Speaker(s): Kazuhiko Hayakawa

In this paper, we propose a new approach to estimate short panel regression models. The model considered is general enough: the model can be either static or dynamic, the time-varying regressors can include endogenous, predetermined and/or strictly exogenous variables, and time invariant regressors can also be included. The errors can contain standard fixed effect and/or interactive fixed effects. Thereby, our setup includes a model with simultaneity and/or a measurement error implicitly.  We propose the maximum likelihood(ML) and minimum distance(MD) estimators to estimate these models in a unified way by utilizing the covariance structure analysis. The distinct feature of our approach is that we do not need to use instrumental variables even in the presence of endogenous regressors. Theoretical investigation demonstrates that a rank deficiency problem arises when there is an endogenous regressor, and a simple solution to address that problem is proposed. Monte Carlo simulation results reveals that the proposed ML and MD estimators outperform most of the existing estimators such as GMM estimators.

Location: Staff Room - A/EC202

Admission: Staff and PhD