The principal objective of this project is to extend the theoretical and empirical knowledge on the interaction of temporal aggregation and non-stationary seasonality and their consequences for macro-econometric modeling. Specifically , the seven main objectives are as follows:
(1) To analyze the effects of different schemes of sampling when the univariate representation measured at weekly monthly or quarterly basis, contains unit roots at their seasonal frequencies.
(2) To investigate the empirical relevance of the aliasing effect in principal European macroeconomic time series.
(3) To study the effects of systematic average, and mixed sampling (when the time series in a system are temporally aggregated by different schemes) on co integrated systems where some or all the variables can contain seasonal unit roots in their univariate representations.
(4) To build up a small empirical seasonal co integrated model at different frequency of measurement of the time series, i.e. monthly, quarterly and semi-annually.
(5) To more rigorously investigate the effects of temporal aggregation on the behavior of unit root tests when seasonal unit roots are present in the data generating process.
(6) To analyze the effects of different sampling schemes on tests of co integration.
(7) To study the effects of temporal aggregation on the estimation and hypothesis testing of co integrating.