A1498
Title: Almost unbiased variance estimation in simultaneous equation models
Authors: Garry Phillips - Cardiff University (United Kingdom)
Yongdeng Xu - Cardiff University (United Kingdom)
Yongdeng Xu - Cardiff University (United Kingdom) [presenting]
Abstract: While a good deal of research in simultaneous equation models has been conducted to examine the small sample properties of coefficient estimators, there has not been a corresponding interest in the properties of estimators for the associated variances. Building on a past study, the biases are explored in variance estimators. This is done for the 2SLS and the MLIML estimators. The approximations to the bias are then used to develop less biased estimators whose properties are examined and compared in a number of simulation experiments. The experiments also consider coverage probabilities/test sizes and test powers of the t-tests, where it is shown that tests based on 2SLS are generally oversized. In contrast, test sizes based on MLIML are closer to nominal levels. In both cases, test statistics based on the corrected variance estimates generally have a higher power than standard procedures. The practical relevance is illustrated in a few well-known applications.
