Some exact distribution theory for maximum likelihood estimators of cointegrating coefficients

This paper derives some exact finite sample distributions and characterizes the tail behavior of maximum likelihood estimators of the cointegrating coefficients in error correction models. It is shown that the reduced rank regression estimator has a distribution with Cauchy-like tails and no finite moments of integer order. The maximum likelihood estimator of the coefficients in a particular triangular system representation is studied and shown to have matrix t-distribution tails with finite integer moments to order T - n + r where T is the sample size, n is the total number of variables in the system and r is the dimension of the cointegration space. These results help to explain some recent simulation studies where extreme outliers are found to occur more frequently for the reduced rank regression estimator than for alternative asymptotically efficient procedures that are based on the triangular representation. In a simple triangular system, the Wald statistic for testing linear hypotheses about the columns of the cointegrating matrix is shown to have an F distribution, analogous to Hotelling's T^{2} distribution in multivariate linear regression.

[1]  C. Herz.  BESSEL FUNCTIONS OF MATRIX ARGUMENT , 1955 .

[2]  Peter C. B. Phillips,et al.  Spectral Regression for Cointegrated Time Series , 1988 .

[3]  Peter C. B. Phillips,et al.  Statistical Inference in Instrumental Variables Regression with I(1) Processes , 1990 .

[4]  P. Phillips,et al.  Vector autoregression and causality: a theoretical overview and simulation study , 1994 .

[5]  P. Phillips,et al.  Statistical Inference in Instrumental Variables , 1989 .

[6]  P. Phillips.  Understanding spurious regressions in econometrics , 1986 .

[7]  A. James.  Normal Multivariate Analysis and the Orthogonal Group , 1954 .

[8]  P. Phillips.  The Exact Distribution of LIML: II , 1984 .

[9]  P. Phillips.  Fractional Matrix Calculus and the Distribution of Multivariate Tests , 1987 .

[10]  R. Muirhead.  Aspects of Multivariate Statistical Theory , 1982, Wiley Series in Probability and Statistics.

[11]  Gregory C. Reinsel,et al.  Nested Reduced-Rank Autoregressive Models for Multiple Time Series , 1988 .

[12]  R. Muirhead,et al.  Asymptotic expansions for distributions of latent roots in multivariate analysis , 1976 .

[13]  The Distribution of FIML in the Leading Case , 1986 .

[14]  Peter C. B. Phillips,et al.  Exact Small Sample Theory in the Simultaneous Equations Model , 1983 .

[15]  S. Johansen.  Estimation and Hypothesis Testing of Cointegration Vectors in Gaussian Vector Autoregressive Models , 1991 .

[16]  Peter C. B. Phillips,et al.  Estimating Long Run Economic Equilibria , 1991 .

[17]  Mark W. Watson,et al.  A SIMPLE ESTIMATOR OF COINTEGRATING VECTORS IN HIGHER ORDER INTEGRATED SYSTEMS , 1993 .

[18]  S. Johansen.  STATISTICAL ANALYSIS OF COINTEGRATION VECTORS , 1988 .

[19]  P. Phillips.  Spherical matrix distributions and cauchy quotients , 1989 .

[20]  Peter C. B. Phillips,et al.  Reflections on econometric methodology , 1988 .

[21]  Joon Y. Park.  Canonical Cointegrating Regressions , 1992 .

[22]  Peter C. B. Phillips,et al.  Optimal Inference in Cointegrated Systems , 1991 .

[23]  Pentti Saikkonen,et al.  Asymptotically Efficient Estimation of Cointegration Regressions , 1991, Econometric Theory.

[24]  G. Hillier,et al.  ON THE NORMALIZATION OF STRUCTURAL EQUATIONS: PROPERTIES OF DIRECTION ESTIMATORS' , 1990 .

[25]  M. Ogaki,et al.  Inference in Cointegrated Models Using VAR Prewhitening to Estimate Shortrun Dynamics , 1991 .

[26]  Allan W. Gregory.  Testing for Cointegration in Linear Quadratic Models , 1994 .

[27]  P. Phillips.  The Exact Distribution of the Wald Statistic , 1986 .