In this paper, we propose a new method for estimating and forecasting asymmetric stochastic volatility models. The proposal is based on dynamic linear models with Markov switching written as state space models. Then, the likelihood is calculated through Kalman filter outputs and the estimates are obtained by the maximum likelihood method. Monte Carlo experiments are performed to assess the quality of estimation. In addition, a backtesting exercise with the real-life time series illustrates that the proposed method is a quick and accurate alternative for forecasting value-at-risk.
The paper assesses the method proposed by Shumway and Stoffer (2006, Chapter 6, Section 10) to estimate the parameters and volatility of stochastic volatility models. First, the paper presents a Monte Carlo evaluation of the parameter estimates considering several distributions for the perturbations in the observation equation. Second, the method is assessed empirically, through backtesting evaluation of VaR forecasts of the S&P 500 time series returns. In both analyses, the paper also evaluates the convenience of using the Fuller transformation.
3 April 2017 in Communications in Statistics: Case Studies, Data Analysis and Applications
In this work, we applied pair-copula models to estimate the market risk of a portfolio composed by future contracts. Pair-copula models (also known as vine copulas) have received much attention mainly because of their flexibility to reproduce various patterns of correlations and tail dependence, and we assessed how different pair-copula specifications change the risk estimation and decomposition of the chosen portfolio. We conclude that even though the backtesting results are nearly the same, different pair-copula models change the risk decomposition of a portfolio.
6 March 2022 in Studies in Nonlinear Dynamics & Econometrics
Stochastic Volatility (SV) models are an alternative to GARCH models for estimating volatility and several empirical studies have indicated that volatility exhibits long-memory behavior. The main objective of this work is to propose a new method to estimate a univariate long-memory stochastic volatility (LMSV) model. For this purpose we formulate the LMSV model in a state-space representation with non-Gaussian perturbations in the observation equation, and the estimation of parameters is performed by maximizing the likelihood written in terms derived from a Kalman filter algorithm. We also present a procedure to calculate volatility and Value-at-Risks forecasts. The proposal is evaluated by means of Monte Carlo experiments and applied to real-life time series, where an illustration of market risk calculation is presented.