A Simple Estimator of Cointegrating Vectors in Higher Order Integrated Systems (Q156121)

From MaRDI portal

!

This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:

scientific article; zbMATH DE number 447127
Language Label Description Also known as
default for all languages
No label defined
    English
    A Simple Estimator of Cointegrating Vectors in Higher Order Integrated Systems
    scientific article; zbMATH DE number 447127

      Statements

      0 references
      61
      0 references
      4
      0 references
      783
      0 references
      July 1993
      0 references
      30 November 1994
      0 references
      0 references
      0 references
      A Simple Estimator of Cointegrating Vectors in Higher Order Integrated Systems (English)
      0 references
      estimators of cointegrating vectors
      0 references
      cointegrating regressions
      0 references
      Gaussian MLE's
      0 references
      triangular representation
      0 references
      random mixture of normals
      0 references
      Wald test statistics
      0 references
      asymptotic chi-squared null distributions
      0 references
      least squares
      0 references
      finite sample performance
      0 references
      Monte Carlo experiments
      0 references
      dynamic OLS estimator
      0 references
      long-run money demand
      0 references
      income elasticity
      0 references
      interest rate semi- elasticity
      0 references
      James H. Stock
      0 references
      Mark W. Watson
      0 references
      This paper makes three main contributions. First, it develops two alternative, computationally simple estimators of cointegrating vectors. These estimators, which have been independently proposed and studied elsewhere in the case of \(I(1)\) variates, are developed here for cointegrating regressions among general \(I(d)\) variables with general deterministic components. The estimators are motivated as Gaussian MLE's for a particular parametrization of the triangular representation. However, under more general conditions they are asymptotically efficient in \textit{P. Saikkonen} [Econ. Theory 7, 1-21 (1991)] and \textit{P. C. B. Phillips'} [Econometrika 59, No. 2, 283-306 (1991; Zbl 0729.62104)] sense, having an asymptotic distribution that is a random mixture of normals and producing Wald test statistics with asymptotic chi-squared null distributions. In the \(I(1)\) case with a single cointegrating vector, one simply regresses one of the variables onto contemporaneous levels of the remaining variables, leads and lags of their first differences, and a constant, using either ordinary or generalized least squares. The resulting ``dynamic OLS'' (respectively GLS) estimators are asymptotically equivalent to the Johansen/Ahn-Reinsel estimator. The second contribution is an examination of the finite sample performance of these estimators in a variety of Monte Carlo experiments. Although all the estimators performs well when the designs incorporate simple short run dynamics, for designs that mimic the dynamics in U.S. real money (M1) balances, income, and interest rates, there is considerable variation across the estimators and associated confidence intervals. In these designs, the dynamic OLS estimator performs well relative to the other asymptotically efficient estimators. The third contribution is the use of these procedures to investigate the long-run demand for money (M1) in the U.S. from 1900 to 1989. Other researchers have argued either explicitly or implicitly that long-run money demand can be thought of as a cointegrating relation among real balances, real income, and an interest rate in postwar data. The authors find this characterization empirically plausible for the longer annual data as well, and therefore use these estimators of cointegrating vectors to examine {[\textit R. E. Lucas'} [Carnegie-Rochester Conf. Ser. Public Policy 29, 137-168 (1988)] suggestion that there is a stable long-run M1 money demand relation spanning the twentieth century. Based on a full sample, 95\% confidence intervals for the income elasticity and interest rate semi-elasticity are, respectively, (0.88, 1.06) and (\(-0.127\), \(- 0.075\)). The postwar data are dominated by a single long-run trend, reflected in growth in income and interest rates and stable real money balances from 1946 to 1982. This results in imprecise estimation of the long-run money demand parameters when only the postwar data are used.
      0 references

      Identifiers

      0 references
      0 references
      0 references
      0 references
      0 references
      0 references