The first step of the time series analysis is to investigate the stationarity of the series. It is
well known that the traditional unit root tests, based on Dickey and Fuller (1979, 1981) and
Phillips and Perron (1988), present important size and power distortions. However, these
problems have been overcome by important modifications in the test procedure, as
proposed by Perron and Ng (1996), Elliott et al. (1996) and Ng and Perron (2001).As
summarized by Ng and Perron (2002), the proper implementation of the new tests, based on
the DF and Z class of tests, can generate important size and power gains.
months. Finally, the interest rate (it), represented by the call rate, was measured in the
correspondent unit of time, given by the accumulated rate over the last 6 months. It is
important to notice that this formulation makes use of backward looking expectations.
2
All
the variables were obtained from the International Financial Statistics (IMF).
The first step of the time series analysis is to investigate the stationarity of the series. It is
well known that the traditional unit root tests, based on Dickey and Fuller (1979, 1981) and
Phillips and Perron (1988), present important size and power distortions. However, these
problems have been overcome by important modifications in the test procedure, as
proposed by Perron and Ng (1996), Elliott et al. (1996) and Ng and Perron (2001).As
summarized by Ng and Perron (2002), the proper implementation of the new tests, based on
the DF and Z class of tests, can generate important size and power gains.
months. Finally, the interest rate (it), represented by the call rate, was measured in the
correspondent unit of time, given by the accumulated rate over the last 6 months. It is
important to notice that this formulation makes use of backward looking expectations.
2
All
the variables were obtained from the International Financial Statistics (IMF).
การแปล กรุณารอสักครู่..
