Persistence detection for the application of a complex algorithm in volatility prediction

Authors

Briones Zúñiga, José Luis

Arancel Llerena, Felipe Fernando

Abstract

The research studies the dynamics of the volatility of the nominal exchange rate of the Peruvian nuevo sol against the US dollar, with the aim of identifying signs of dependencies over long time distances. The existence of persistence in yields and their volatility is explored by applying two tests based on periograms, the Hurst exponent is estimated, its consistency over time and the graphical analysis of its distributions for the subsequent application of the complex dynamics algorithm. For this objective, the time series of the nominal exchange rate of the nuevo sol against the dollar from January 3, 2013 to February 11, 2021 is used. It is concluded that there is sufficient evidence of a non-random fractal pattern conducive to the use of the algorithm to propose.