|
|
|
| A New Method of Detecting Nonlinear for Time Series Based on KS Test |
| Hou Shu-min; Li You-rong; Liu Guang-ling |
| College of Mechanical Engineering, Wuhan University of Science and Technology, Wuhan 430081, China |
|
|
|
|
Abstract The choice of test statistics can bring important influence to nonlinear of time series. This paper introduces a Non-parameter test——Kolmogorov-Smirnov (KS) test into nonlinearity test. After applying the Phase-randomized surrogate algorithm to create surrogate data, three test statistics methods, which include KS test, the third-order autocovariance and the asymmetry due to time reversal, are employed to determine nonlinear of five kinds signals. By comparing the test results, it indicates that KS test is an effective and stable nonlinear test statistics. The proposed method has high noise immunity and more sensitive to nonlinear signal.
|
|
Received: 12 August 2005
|
|
|
|
|
|
|
|
|
2007, Vol. 29 
