Higher-Order aymptotic expansions of the least-squares estimation bias in first-order dynamic regression models

Kiviet, Jan and Phillips, Garry David Alan 2012. Higher-Order aymptotic expansions of the least-squares estimation bias in first-order dynamic regression models. Journal of Computational Statistics and Data Analysis 56 (11) , pp. 3706-3729. 10.1016/j.csda.2010.07.013

Full text not available from this repository.

Abstract

An approximation to orderT−2 is obtained for the bias of the full vector of least-squares estimates obtained from a sample of size T in general stable but not necessarily stationary ARX(1) models with normal disturbances. This yields generalizations, allowing for various forms of initial conditions, of Kendall’s and White’s classic results for stationary AR(1) models. The accuracy of various alternative approximations is examined and compared by simulation for particular parameterizations of AR(1) and ARX(1) models. The results show that often the second-order approximation is considerably better than its first-order counterpart and hence opens up perspectives for improved bias correction. However, orderT−2 approximations are also found to be more vulnerable in the near unit root case than the much simpler orderT−1 approximations.

Item Type:	Article
Date Type:	Publication
Status:	Published
Schools:	Schools > Business (Including Economics)
Subjects:	H Social Sciences > HB Economic Theory Q Science > QA Mathematics
Uncontrolled Keywords:	ARX-model; Asymptotic expansion; Bias approximation; Lagged dependent variable; Monte Carlo simulation
Publisher:	Elsevier
Last Modified:	07 Feb 2022 11:12
URI:	https://orca.cardiff.ac.uk/id/eprint/33364

Citation Data

Cited 11 times in Scopus. View in Scopus. Powered By Scopus® Data

Actions (repository staff only)

Edit Item