Modelización y Pronóstico de la Volatilidad del Mercado de Valores Ecuatoriano: Una Aplicación de Modelos GARCH

Abstract

Abstract: This research aims to model and forecast the volatility of the Ecuadorian stock market using the Ecuindex stock market index by applying generalized autoregressive conditional heteroskedasticity (GARCH) models. Understanding volatility in emerging markets such as Ecuador is essential for risk management and informed financial decision-making. Daily data from the Ecuindex index, corresponding to 4,254 observations, were used for the analysis. The process began with a statistical and graphical exploration of returns, which identified characteristics specific to financial series, such as volatility clustering and conditional heteroskedasticity. Various GARCH model specifications were then estimated, including GARCH(1,1), GARCH(1,2), and GARCH(2,1). The selection of the optimal model was based on the Akaike Information Criterion (AIC) and the statistical significance of the parameters. The chosen model was a GARCH(1,3), which presented a good fit and relevant parameters. Diagnostic tests confirmed the absence of remaining heteroskedasticity, validating its suitability and allowing for the generation of volatility forecasts for economic agents operating in the Ecuadorian market.