Análise estatística da volatilidade em Petr4 e Vale3 com o Modelo Aparch: uma perspectiva sobre períodos de crise

Abstract

Risk and uncertainty quantification are crucial aspects of the financial sector of an economy. From the perspective of financial institutions, risk analysis enables the development of better strategies for asset portfolio management and resource allocation. The objective of this study is to examine clustering, persistence, and asymmetry in the volatility of Petrobras (Petróleo Brasileiro S/A), a mixed-capital company controlled by the Federal Government, and Vale do Rio Doce, a privately controlled institution, both listed on the Brazilian stock exchange (B3), during periods of economic or political instability in the national scenario. In an initial approach, the time series of daily closing prices of Petrobras' preferred shares, symbolized by PETR4, and Vale’s common shares, symbolized by VALE3, were considered for the period from January 2007 to December 2024. Volatility modeling was performed using the Asymmetric Power Autoregressive Conditional Heteroscedasticity (APARCH) model, allowing the evaluation of asymmetry dynamics and the persistence of shocks in returns throughout the study period. Preliminary results suggest that negative shocks have asymmetrically impacted the price trajectories of these assets compared to positive shocks, as evidenced by a steeper impact curve. This reveals a greater sensitivity of these assets to adverse events, increasing risk during specific periods of instability.