Stochastic models for fractal processes (Q1304354)

The authors establish the situation where a stochastic process may display both long range dependence and intermittency. The spectral density \[ f(\omega)=c \{|\omega|^{2\gamma}(1+\omega^2)^\alpha\}^{-1},\quad c>0,\;1/2<\gamma <3/2,\;\alpha \geq 0,\;\omega \in R, \] is considered. A new approach for estimating the parameters \(\alpha\) and \(\gamma\) simultaneously is proposed. This approach is based on wavelet theory. A simulation study on the performance of this method is given.