| Sumario: | The writer examines nonstationary stochastic seasonality to model the German M2 money demand function using quarterly data for the period spanning 1960-96. He finds empirical evidence in favor of a stable long-run M2 money demand function relying on seasonally unadjusted data, but he points out that it is not possible to establish a stable long-run relationship using seasonally adjusted time series. He contends that the seasonal error correction model exhibits satisfactory properties and fits the data quite well. He points out, however, that it displays parameter nonconstancy, particularly in the middle of the 1970s and at the start of the 1990s, when major monetary regime changes occurred.
|
|---|