ZHOU ZHOU - University of Toronto
The assumption of (weak) stationarity is crucial for the validity of most of the
conventional tests of structure change. Under complicated non-stationary temporal dynamics, we argue that traditional testing procedures result in mixed structural change signals of the first and second order and hence could lead to seriously biased testing results. We proposes a simple and unied bootstrap testing procedure which provides consistent testing results under very general forms of smooth and abrupt changes in the dynamics of the time series.