Long Run Estimations for the Volatility of Time Series in the Brazilian Financial Market

Main Article Content

Alex Sandro Monteiro de Moraes
Antonio Carlos Figueiredo Pinto
Marcelo Cabus Klotzle

Abstract

The models of the GARCH family, normally used for the estimates of volatility for longer periods, keep unchanged the relative weights assigned to the observations both old and new, regardless of the volatility´s forecasted horizon. The purpose of this article is to verify if the increase in relative weights assigned to the earlier observations due to the increase of the forecast horizon results in better estimates of volatility. Through the use of seven forecasting models of volatility and return series of financial markets assets, the estimates obtained in the sample (in-sample) were compared with observations outside the sample (out-of-sample). Based on this comparison, it was found that the best estimates of expected volatility were obtained by the modified EGARCH model and the ARLS model. We conclude that the use of traditional forecasting models of volatility, which keep unchanged relative weights assigned to both old and new observations, was inappropriate.

Article Details

Section
Long Paper
Author Biography

Marcelo Cabus Klotzle, Pontifícia Universidade Católica do Rio de Janeiro

Ph.D em economia pela Katholische Universität Eichstätt (Alemanha) É professor do IAG (Departamento de Administração) da PUC-Rio na área de Finanças e Economia