A multivariate approach to modeling univariate seasonal time series
A seasonal time series can be represented by a vector autoregressive model for the annual series containing the seasonal observations. This model allows for periodically varying coefficients. When the vector elements are integrated, the maximum likelihood cointegration method can be used to check for the presence of, possibly restricted, cointegration relations between these annual series. In this paper it is shown that this application generalizes a test procedure for seasonal unit roots. Simulations and examples illustrate its empirical performance.
|Keywords||cointegration, seasonality, time series, vector autoregression|
|JEL||Time-Series Models; Dynamic Quantile Regressions (jel C22)|
|Persistent URL||dx.doi.org/10.1016/0304-4076(93)01563-2, hdl.handle.net/1765/2079|
|Journal||Journal of Econometrics|
Franses, Ph.H.B.F. (1994). A multivariate approach to modeling univariate seasonal time series. Journal of Econometrics, 133–151. doi:10.1016/0304-4076(93)01563-2