A time-domain test for some types of nonlinearity

Abstract

Abstract-The bispectrum and third-order moment can be viewed as equivalent tools for testing for the presence of nonlinearity in stationary time series. This is because the bispectrum is the Fourier transform of the third-order moment. An advantage of the bispectrum is that its estimator comprises terms that are asymptotically independent at distinct bifrequencies under the null hypothesis of linearity. An advantage of the third-order moment is that its values in any subset of joint lags can be used in the test, whereas when using the bispectrum the entire (or truncated) third-order moment is required to construct the Fourier transform. In this paper, we propose a test for nonlinearity based upon the estimated third-order moment. We use the phase scrambling bootstrap method to give a nonparametric estimate of the variance of our test statistic under the null hypothesis. Using a simulation study, we demonstrate that the test obtains its target significance level, with large power, when compared to an existing standard parametric test that uses the bispectrum. Further we show how the proposed test can be used to identify the source of nonlinearity due to interactions at specific frequencies. We also investigate implications for heuristic diagnosis of nonstationarity.

Predictability of Marine Population Trajectories–the effect of delays and resource availability

“…measure of symmetry around the mean of the time series, of the original with the surrogate time series based on test statistics, see [31,34,35].…”

“…The bootstrapping method from [34], aims to find a third-order estimate in the region 0 ≤ ξ 1 ≤ ξ 2 ≤ M , with M as a truncation value (1 ≤ M ≤ N ). The integers ξ 1 and ξ 2 define the serial indicates for the calculation of the third-order moments, i.e.…”

A bootstrap test for time series linearity

“…There are many published reports, especially in recent years, that utilize the bootstrap in linearity testing, e.g., Barnett and Wolff (2005), Birkelund and Hanssen (2009), Hinich et al (2005), Hjellvik and Tjostheim (1995), Kugiumtzis (2008). Our approach is fundamentally different from these previous works although some similarities are present.…”

Harmonic Regression Models: A Comparative Review with Applications

“…The typical null being that the series under consideration is both linear and stationary, a rejection could be either in the direction of nonlinearity, nonstationarity or both. Recent attempts to separate nonlinearity and nonstationarity via so-called phase scrambled bootstrap methods (Aparicio (1998), Kugimutzis (2001) and Barnett and Wolff (2005)) have not been very successful. Method V attempts to meet this challenging task by attempting to model both the nonlinear and possibly nonstationary features of the data simultaneously, by making the coefficients time varying in a deterministic fashion.…”