We develop an asymptotically chi-squared test statistic for testing moment conditions E[mt( 0 )] = 0 where scalar components of mt( 0 ) may have an in…nite variance and mt( 0 ) may be weakly dependent. In general E[mt( 0 )] need not exist under the alternative. A variety of tests can be heavy-tail robusti…ed by our method, including white noise, GARCH a¤ects, omitted variables, order selection, functional form, causation, volatility spillover and over-identi…cation. The test statistic is derived from a tail-trimmed sample version of the moments evaluated at a consistent plug-in^ T for 0 . Depending on the test in question^ T may be any consistent estimator like QML, LAD, GMM, and Empirical Likelihood as well as robust estimators like Least Trimmed Squares, Least Absolute Weighted Deviations, and Generalized Method of Tail-Trimmed Moments. Simple rules of thumb for selecting the trimming fractiles are presented, and in many cases when mt( 0 ) has in…nite variance components the fractiles and/or^ T can be chosen to ensure^ T does not in ‡uence the test statistic's limit distribution. Thus, in heavy tailed cases^ T does not need to have a Gaussian limit. We apply our statistic to tests of white noise, omitted variables and volatility spillover and …nd it obtains correct empirical size, while conventional tests exhibit sharp distortions.
We study the dynamics of pricing efficiency in the equity REIT market from 1993 to 2014. We measure pricing efficiency at the firm level using variance ratios calculated from quote midpoints in the TAQ database. We find four main results. First, on average, the market is efficient, with variance ratios close to one. However, in any given year, there is considerable cross‐sectional variation in variance ratios, suggesting at least some firms are priced inefficiently. Second, higher institutional ownership by active institutional investors is related to better pricing efficiency, while passive ownership does not reduce pricing efficiency. Third, REITs that are included in the S&P 500 and S&P 400 are priced more efficiently than other REITs. For the S&P 500 firms, we find evidence that this was purely driven by sample selection, while for S&P 400 firms, we find evidence that it is inclusion in the index that drives efficiency. Finally, we find evidence that firm investment, analyst coverage and debt capital raising activity can influence pricing efficiency.
We develop an asymptotically chi-squared test statistic for testing moment conditions E[mt( 0 )] = 0 where scalar components of mt( 0 ) may have an in…nite variance and mt( 0 ) may be weakly dependent. In general E[mt( 0 )] need not exist under the alternative. A variety of tests can be heavy-tail robusti…ed by our method, including white noise, GARCH a¤ects, omitted variables, order selection, functional form, causation, volatility spillover and over-identi…cation. The test statistic is derived from a tail-trimmed sample version of the moments evaluated at a consistent plug-in^ T for 0 . Depending on the test in question^ T may be any consistent estimator like QML, LAD, GMM, and Empirical Likelihood as well as robust estimators like Least Trimmed Squares, Least Absolute Weighted Deviations, and Generalized Method of Tail-Trimmed Moments. Simple rules of thumb for selecting the trimming fractiles are presented, and in many cases when mt( 0 ) has in…nite variance components the fractiles and/or^ T can be chosen to ensure^ T does not in ‡uence the test statistic's limit distribution. Thus, in heavy tailed cases^ T does not need to have a Gaussian limit. We apply our statistic to tests of white noise, omitted variables and volatility spillover and …nd it obtains correct empirical size, while conventional tests exhibit sharp distortions.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.