A WAVELET METHOD COUPLED WITH QUASI-SELF-SIMILAR STOCHASTIC PROCESSES FOR TIME SERIES APPROXIMATION

Mohamed Essaied Hamrita, Nidhal Ben Abdallah, Anouar Ben Mabrouk

International Journal of Wavelets Multiresolution and Information Processing · 2011 · 22 citations · 18 references

Concepts

Abstract

Scaling laws and generally self-similar structures are now well known facts in financial time series. Furthermore, these signals are characterized by the presence of stochastic behavior allowing their analysis with pure functional methods being incomplete. In the present paper, some existing models are reviewed and modified, based on wavelet theory and self-similarity, to recover multi-scaling cases for approximating financial signals. The resulting models are then tested on some empirical examples and analyzed for error estimates.

References

18