Statistical inference in a random coefficient panel model

This paper studies the asymptotics of the Weighted Least Squares (WLS) estimator of the autoregressive root in a panel Random Coefficient Autoregression (RCA). We show that, in an RCA context, there is no “unit root problem” : the WLS estimator is always asymptotically normal, irrespective of the av...

Full description

Bibliographic Details
Main Authors: Horváth, Lajos, Trapani, Lorenzo
Format: Article
Published: Elsevier 2016
Subjects:
Online Access:https://eprints.nottingham.ac.uk/46953/
_version_ 1848797435731640320
author Horváth, Lajos
Trapani, Lorenzo
author_facet Horváth, Lajos
Trapani, Lorenzo
author_sort Horváth, Lajos
building Nottingham Research Data Repository
collection Online Access
description This paper studies the asymptotics of the Weighted Least Squares (WLS) estimator of the autoregressive root in a panel Random Coefficient Autoregression (RCA). We show that, in an RCA context, there is no “unit root problem” : the WLS estimator is always asymptotically normal, irrespective of the average value of the autoregressive root, of whether the autoregressive coefficient is random or not, and of the presence and degree of cross dependence. Our simulations indicate that the estimator has good properties, and that confidence intervals have the correct coverage even for sample sizes as small as (N,T)=(10,25)(N,T)=(10,25). We illustrate our findings through two applications to macroeconomic and financial variables.
first_indexed 2025-11-14T20:03:50Z
format Article
id nottingham-46953
institution University of Nottingham Malaysia Campus
institution_category Local University
last_indexed 2025-11-14T20:03:50Z
publishDate 2016
publisher Elsevier
recordtype eprints
repository_type Digital Repository
spelling nottingham-469532020-05-04T20:02:05Z https://eprints.nottingham.ac.uk/46953/ Statistical inference in a random coefficient panel model Horváth, Lajos Trapani, Lorenzo This paper studies the asymptotics of the Weighted Least Squares (WLS) estimator of the autoregressive root in a panel Random Coefficient Autoregression (RCA). We show that, in an RCA context, there is no “unit root problem” : the WLS estimator is always asymptotically normal, irrespective of the average value of the autoregressive root, of whether the autoregressive coefficient is random or not, and of the presence and degree of cross dependence. Our simulations indicate that the estimator has good properties, and that confidence intervals have the correct coverage even for sample sizes as small as (N,T)=(10,25)(N,T)=(10,25). We illustrate our findings through two applications to macroeconomic and financial variables. Elsevier 2016-07 Article PeerReviewed Horváth, Lajos and Trapani, Lorenzo (2016) Statistical inference in a random coefficient panel model. Journal of Econometrics, 193 (1). pp. 54-75. ISSN 0304-4076 Random Coefficient Autoregression; Panel data; WLS estimator; Common factors http://www.sciencedirect.com/science/article/pii/S0304407616300203 doi:10.1016/j.jeconom.2016.01.006 doi:10.1016/j.jeconom.2016.01.006
spellingShingle Random Coefficient Autoregression; Panel data; WLS estimator; Common factors
Horváth, Lajos
Trapani, Lorenzo
Statistical inference in a random coefficient panel model
title Statistical inference in a random coefficient panel model
title_full Statistical inference in a random coefficient panel model
title_fullStr Statistical inference in a random coefficient panel model
title_full_unstemmed Statistical inference in a random coefficient panel model
title_short Statistical inference in a random coefficient panel model
title_sort statistical inference in a random coefficient panel model
topic Random Coefficient Autoregression; Panel data; WLS estimator; Common factors
url https://eprints.nottingham.ac.uk/46953/
https://eprints.nottingham.ac.uk/46953/
https://eprints.nottingham.ac.uk/46953/