Forecasting Volatility with Copula-Based Time Series Models

Abstract

Standard-Nutzungsbedingungen:Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden.Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen.Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Abstract This paper develops a novel approach to modeling and forecasting realized volatility (RV) measures based on copula functions. Copula-based time series models can capture relevant characteristics of volatility such as nonlinear dynamics and long-memory type behavior in a flexible yet parsimonious way. In an empirical application to daily volatility for S&P500 index futures, we find that the copula-based RV (C-RV) model outperforms conventional forecasting approaches for one-day ahead volatility forecasts in terms of accuracy and efficiency. Among the copula specifications considered, the Gumbel C-RV model achieves the best forecast performance, which highlights the importance of asymmetry and upper tail dependence for modeling volatility dynamics. Although we find substantial variation in the copula parameter estimates over time, conditional copulas do not improve the accuracy of volatility forecasts. Terms of use: Documents in

Forecasting time series with multivariate copulas

“…In this paper, we extend the methodology of [26] by proposing a forecasting method using copula-based models for multivariate time series, as in [23]. As one can guess, we show that forecasting multivariate time series using copula-based models gives better results than forecasting a single time series, since more information means more precision, in general.…”

“…As mentioned previously, we are going to compare our predictions performance with the univariate version presented in [26], when (X ,t ) is a Markov process. In this case, let D be the copula associated with (X ,t− , X ,t ) for t ∈ N, i.e., D is the copula of (U ,t− , U ,t ).…”

“…Here we compare the performance of our forecasting method versus the method proposed by [26]. The analysis is restricted to bivariate time series, but the results can easily be extrapolated to higher dimensions.…”

“…Once again, the exibility of copulas allows to model more complex dependence structures and thus to better capture the evolution of the time series. In the recent work of [26], copulas were used to forecast the realized volatility associated with a univariate nancial time series and they showed that copula-based forecasts perform better than forecasts based on heterogeneous autoregressive (HAR) model. The later method had been proven successful in [3], [10] and [8].…”

Effects of Political Instability on the Volatility of Palestinian Food Prices

“…Enriching the analysis with political and trade aspects would be valuable too which depends on sufficiently available data. Last, carrying out a comparative analysis of the performance of models for improving forecasts of food price volatility, e.g., using copula-based approaches (Sokolinskiy and van Dijk, 2011;Patton, 2012), would create significant insight for stakeholders.…”