J. Timmer, POWER OF SURROGATE DATA TESTING WITH RESPECT TO NONSTATIONARITY, Physical review. E, Statistical physics, plasmas, fluids, and related interdisciplinary topics, 58(4), 1998, pp. 5153-5156
Surrogate data testing is a method frequently applied to evaluate the
results of nonlinear time series analysis. Since the null hypothesis t
ested against is a linear, Gaussian, stationary stochastic process a p
ositive outcome may not only result from an underlying nonlinear or ev
en chaotic system, but also from, e.g., a nonstationary linear one. We
investigate the power of the test against nonstationarity. [S1063-651
X(98)14610-X].