Testing the joint hypothesis of rationality and neutrality under seasonal cointegration: The case of Korea

With seasonally adjusted data, and using a procedure based on nonseasonal cointegration, the macro rational expectations hypothesis of rationality and money neutrality is rejected at the 10% level. However, with seasonally unadjusted data, and using a procedure based on seasonal cointegration, the same hypothesis is not rejected. The paper provides an example of how deseasonalizing variable by variable can distort empirical inference in two important ways: by introducing noninvertibility at the seasonal frequencies or by failing to take into account the presence of cointegrating relations at these frequencies.