An autoregressive distributed-lag modelling approach to cointegration analysis.

*(English)*Zbl 0972.91076
StrĂ¸om, Steinar (ed.), Econometrics and economic theory in the 20th century. The Ragnar Frisch centennial symposium. Papers from the symposium held in Oslo, Norway, March 3-5, 1995. Cambridge: Cambridge University Press. Econ. Soc. Monogr. 31, 371-413 (1998).

Econometric analysis of long-run relations has been the focus of much theoretical and empirical research in economics. In cases where the variables in the long-run relation of interest are trend-stationary, the general practice has been to de-trend the series and two model the de-trended series as stationary autoregressive distributed-lag models. Estimation and inference concerning the long-run properties of the model have then been carried out using standard asymptotic normal theory. The analysis becomes more complicated when the variables are difference-stationary or integrated of order 1. The recent literature on cointegration has been concerned analysis of the long-run relations between such variables, and its basic premise has been, at least implicitly, that in the presence of such variables, the traditional autoregressive distributed-lag approach is no longer applicable.

In this chapter , the authors re-examine the use of the traditional autoregressive distributed-lag approach for the analysis of long-run relations when the underlying variables are difference-stationary or integrated of order 1. The presented theoretical analysis and the Monte Carlo simulation results provide strong evidence in favour of a rehabilitation of the traditional autoregressive distributed-lag approach to time-series econometric modelling. The focus, however, has been exclusively on single-equation estimation techniques, while system estimation has not been addressed.

For the entire collection see [Zbl 0944.00092].

In this chapter , the authors re-examine the use of the traditional autoregressive distributed-lag approach for the analysis of long-run relations when the underlying variables are difference-stationary or integrated of order 1. The presented theoretical analysis and the Monte Carlo simulation results provide strong evidence in favour of a rehabilitation of the traditional autoregressive distributed-lag approach to time-series econometric modelling. The focus, however, has been exclusively on single-equation estimation techniques, while system estimation has not been addressed.

For the entire collection see [Zbl 0944.00092].

Reviewer: Vangelis Grigoroudis (Chania)