Robinson, Peter M. and Gerolimetto, M. (2006) Instrumental variables estimation of stationary and nonstationary cointegrating regressions. EM/2006/500. Suntory and Toyota International Centres for Economics and Related Disciplines, London School of Economics and Political Science, London, UK.
Download (354Kb) | Preview
Instrumental variables estimation is classically employed to avoid simultaneous equations bias in a stable environment. Here we use it to improve upon ordinary least squares estimation of cointegrating regressions between nonstationary and/or long memory stationary variables where the integration orders of regressor and disturbance sum to less than 1, as happens always for stationary regressors, and sometimes for mean-reverting nonstationary ones. Unlike in the classical situation, instruments can be correlated with disturbances and/or uncorrelated with regressors. The approach can also be used in traditional non-fractional cointegrating relations. Various choices of instrument are proposed. Finite sample performance is examined.
|Item Type:||Monograph (Discussion Paper)|
|Additional Information:||© 2006 the author|
|Uncontrolled Keywords:||Cointegration; Instrumental variables estimation; I (d) processes|
|Library of Congress subject classification:||H Social Sciences > HB Economic Theory|
|Journal of Economic Literature Classification System:||C - Mathematical and Quantitative Methods > C3 - Econometric Methods: Multiple; Simultaneous Equation Models; Multiple Variables; Endogenous Regressors > C32 - Time-Series Models|
|Sets:||Collections > Economists Online
Departments > Economics
Research centres and groups > Suntory and Toyota International Centres for Economics and Related Disciplines (STICERD)
Actions (login required)
|Record administration - authorised staff only|