Seasonality Tests

This article modifies and extends the test against nonstationary stochastic seasonality proposed by Canova and Hansen. A simplified form of the test statistic in which the nonparametric correction for serial correlation is based on estimates of the spectrum at the seasonal frequencies is considered and shown to have the same asymptotic distribution as the original formulation. Under the null hypothesis, the distribution of the seasonality test statistics is not affected by the inclusion of trends, even when modified to allow for structural breaks, or by the inclusion of regressors with nonseasonal unit roots. A parametric version of the test is proposed, and its performance is compared with that of the nonparametric test using Monte Carlo experiments. A test that allows for breaks in the seasonal pattern is then derived. It is shown that its asymptotic distribution is independent of the break point, and its use is illustrated with a series on U.K. marriages. A general test against any form of permanent seasonality, deterministic or stochastic, is suggested and compared with a Wald test for the significance of fixed seasonal dummies. It is noted that tests constructed in a similar way can be used to detect trading-day effects. An appealing feature of the proposed test statistics is that under the null hypothesis, they all have asymptotic distributions belonging to the Cramér–von Mises family.

[1]  J. Stock.  Unit roots, structural breaks and trends , 1986 .

[2]  Eric Ghysels,et al.  Changes in Seasonal Patters: Are They Cyclical , 1994 .

[3]  Tommaso Proietti,et al.  Comparing seasonal components for structural time series models , 2000 .

[4]  Andrew Harvey,et al.  TESTS OF COMMON STOCHASTIC TRENDS , 2000, Econometric Theory.

[5]  J. Bai,et al.  Estimation of a Change Point in Multiple Regression Models , 1997, Review of Economics and Statistics.

[6]  Andrew Harvey,et al.  Testing for the Presence of a Random Walk in Series with Structural Breaks , 1998 .

[7]  Mehmet Caner.  A Locally Optimal Seasonal Unit-Root Test , 1998 .

[8]  Pentti Saikkonen,et al.  Testing for a Moving Average Unit Root in Autoregressive Integrated Moving Average Models , 1993 .

[9]  Richard J. Smith,et al.  Additional critical values and asymptotic representations for seasonal unit root tests , 1998 .

[10]  A. M. Robert Taylor,et al.  Testing against stochastic trend and seasonality in the presence of unattended breaks and unit roots , 2003 .

[11]  Jeremy Penzer,et al.  Diagnosing Shocks in Time Series , 1998 .

[12]  Andrew Harvey,et al.  Forecasting, Structural Time Series Models and the Kalman Filter , 1990 .

[13]  D. Andrews.  Heteroskedasticity and Autocorrelation Consistent Covariance Matrix Estimation , 1991 .

[14]  Jurgen A. Doornik,et al.  Statistical algorithms for models in state space using SsfPack 2.2 , 1999 .

[15]  J. Nyblom.  Testing for the Constancy of Parameters over Time , 1989 .

[16]  C. Z. Wei,et al.  Limiting Distributions of Least Squares Estimates of Unstable Autoregressive Processes , 1988 .

[17]  David F. Findley,et al.  ON THE SPECTRUM DIAGNOSTICS USED BY X-12-ARIMA TO INDICATE THE PRESENCE OF TRADING DAY EFFECTS AFTER MODELING OR ADJUSTMENT , 1998 .

[18]  D. A. Pierce.  Seasonal adjustment when both deterministic and stochastic seasonality are present , 1978 .

[19]  James D. Hamilton.  The Daily Market for Federal Funds , 1996, Journal of Political Economy.

[20]  Maxwell L. King,et al.  Locally Best Invariant Tests of the Error Covariance Matrix of the Linear Regression Model , 1985 .

[21]  P. Phillips,et al.  Testing the null hypothesis of stationarity against the alternative of a unit root: How sure are we that economic time series have a unit root? , 1992 .

[22]  Brendan McCabe,et al.  A Consistent Test for a Unit Root , 1994 .