NEW SMALL SAMPLE ESTIMATORS FOR COINTEGRATION REGRESSION - LOW-PASS SPECTRAL FILTER METHOD

被引：3

作者：

LI, YK ^{[1
]}

MADDALA, GS ^{[1
]}

RUSH, M ^{[1
]}

机构：

[1] OHIO STATE UNIV,DEPT ECON,COLUMBUS,OH 43210

来源：

ECONOMICS LETTERS | 1995年 / 47卷 / 02期

关键词：

COINTEGRATION; FILTERING; FULLY MODIFIED LEAST SQUARES; SMALL SAMPLE;

D O I：

10.1016/0165-1765(94)00557-I

中图分类号：

F [经济];

学科分类号：

02 ;

摘要：

In this paper we propose a low-pass spectral filter method to estimate cointegrating vectors. The simulation study shows that the small sample properties of the proposed estimators are quite appealing. For the low signal to noise ratio cases in which fully modified least squares (FMLS) is better than ordinary least squares (OLS), filtered least squares beats OLS and filtered FMLS is dominantly better than the FMLS.

引用

页码：123 / 129

页数：7

共 16 条

[1]

BANERJEE A, 1986, OXFORD B ECON STAT, V48, P253

[2]

Boswijk H Peter, 1989, ESTIMATION TESTING C

[3] COINTEGRATION AND ERROR CORRECTION - REPRESENTATION, ESTIMATION, AND TESTING [J].