Modified sign method for testing the fractality of Gaussian noise
- 29 Downloads
Fractal Gaussian noise is a stationary Gaussian sequence of zero-mean random variables whose sums possess the stochastic self-similarity property. If the random variables are independent, the self-similarity coefficient equals 1/2. The sign criterion for testing the hypothesis that the parameter equals 1/2 against the alternative H ≠ 1/2 is based on counting the sign change rate for elements of the sequence. We propose a modification of the criterion: we count sign change indicators not only for the original random variables but also for random variables formed as sums of consecutive elements. The proof of the asymptotic normality of our statistics under the alternative hypothesis is based on the theorem on the asymptotics of the covariance of sign change indicators for a zero-mean stationary Gaussian sequence with a slowly decaying correlation function.
KeywordsCentral Limit Theorem Information Transmission Asymptotic Normality Gaussian Sequence Fractal Gaussian Noise
Unable to display preview. Download preview PDF.
- 2.Henry, M. and Zaffarony, P., The Long-Range Dependence Paradigm for Macroeconomics and Finance, Theory and Applications of Long-Range Dependence, Doukhan, P., Oppenheim, G., and Taqqu, M.S., Eds., Boston: Birkhauser, 2002, pp. 417–438.Google Scholar
- 5.Yaglom, A.M., Korrelyatsionnaya teoriya statsionarnykh sluchainykh funktsii: s primerami iz meteorologii, Leningrad: Gidrometeoizdat, 1981. Translated under the title Correlation Theory of Stationary and Related Random Functions, New York: Springer, 1987.Google Scholar
- 6.Crownover, R.M., Introduction to Fractals and Chaos, Boston: Jones & Bartlett, 1995. Translated under the title Fraktaly i khaos v dinamicheskikh sistemakh. Osnovy teorii, Moscow: Potmarket, 2000.Google Scholar
- 7.Ibragimov, I.A. and Linnik, Yu.V., Nezavisimye i statsionarno svyazannye velichiny, Moscow: Nauka, 1965. Translated under the title Independent and Stationary Sequences of Random Variables, Groningen: Wolters-Noordhoff, 1971.Google Scholar