Kyung So Im

Efficient estimation of panel data models with strictly exogenous explanatory variables

Abstract With panel data, exogeneity assumptions imply many more moment conditions than standard estimators use. However, many of the moment conditions may be redundant, in the sense that they do not increase efficiency; if so, we may establish the standard estimators’ efficiency. We prove efficiency results for GLS in a model with unrestricted error covariance matrix, and for 3SLS in models where regressors and errors are correlated, such as the Hausman–Taylor model. For models with correlation between regressors and errors, and with unrestricted error covariance structure, we provide a simple estimator based on a GLS generalization of deviations from means.

On the Panel Unit Root Tests Using Nonlinear Instrumental Variables

This paper re-examines the panel unit root tests proposed by Chang (2002). She establishes asymptotic independence of the t-statistics when integrable functions of lagged dependent variable are used as instruments even if the original series are cross sectionally dependent. She claims that her non-linear instrumental variable (NIV) panel unit root test is valid under general error cross correlations for any N (the cross section dimension) as T (the time dimension of the panel) tends to infinity. These results are largely due to her particular choice of the error correlation matrix which results in weak cross section dependence. Also, the asymptotic independence property of the t- statistics disappears when Changs modified instruments are used. Using a common factor model with a sizeable degree of cross section correlations, we show that Changs NIV panel unit root test suffers from gross size distortions, even when N is small relative to T.

Robustifying Glejser test of heteroskedasticity

Verifying that the Glejser test for heteroskedasticity is asymptotically invalid unless the error density is symmetric, this paper proposes a simple modification to make the test robust to asymmetric disturbances. Simulation results demonstrate that the size of the modified test is correct for both symmetric and skewed errors.

More Powerful Unit Root Tests with Non-Normal Errors

This paper extends the Lagrange Multiplier (LM) unit toot tests of Schmidt and Phillips (Oxford Bull. Econ. Stat. 54:257–285, 1992) to utilize information contained in non-normal errors. The new tests adopt the Residual Augmented Least Squares (RALS) estimation procedure of Im and Schmidt (J. Econ. 144:219–233, 2008). This paper complements the work of Im et al. (More powerful unit root tests with nonnormal errors, manuscript, 2012) who adopt the RALS procedure for DF-based tests. We provide the relevant asymptotic distribution and the corresponding critical values of our new tests. The RALS-LM tests show improved power over the RALS-DF tests. The main advantage of the RALS-LM tests lies in the invariance feature that the distribution does not depend on the nuisance parameter in the presence of level-breaks.

Panel LM Unit Root Tests with Trend Shifts

This paper proposes a new Lagrange multiplier (LM) based unit root test for panel data allowing for heterogeneous structural breaks in both the intercept and slope of each cross-section unit in the panel. We note that panel unit root tests allowing for breaks in the slope will critically depend on the nuisance parameters indicating the size and location of breaks. Any panel tests that ignore this dependency on the nuisance parameter will be subject to serious size distortions. To address this problem, our test employs a method that renders the asymptotic distribution of the panel tests invariant to nuisance parameters. We derive the asymptotic properties of our test and also examine its finite-sample properties. In addition, our test easily can be modified to correct for the presence of cross-correlations in the innovations of the panel. We illustrate this by applying the cross-sectionally augmented ADF (CADF) procedure of Pesaran (2007) to our test statistic.