Academic literature on the topic 'Multiplicative Component GARCH'

In this paper, we address the issue of how macroeconomic conditions affect corporate bond volatility. We employ the GARCH-MIDAS multiplicative two-component model of volatility that distinguishes the short-term dynamics from the long-run component of volatility. Both the in-sample and out-of-sample analysis show that recognizing the existence of a stochastic low-frequency component captured by macroeconomic and financial indicators may improve the fit of the model to actual bond return data, relative to the constant long-run component embedded in a typical GARCH model.

In this paper, we employ 99% intraday value-at-risk (VaR) and intraday expected shortfall (ES) as risk metrics to assess the competency of the Multiplicative Component Generalised Autoregressive Heteroskedasticity (MC-GARCH) models based on the 1-min EUR/USD exchange rate returns. Five distributional assumptions for the innovation process are used to analyse their effects on the modelling and forecasting performance. The high-frequency volatility models were validated in terms of in-sample fit based on various statistical and graphical tests. A more rigorous validation procedure involves testing the predictive power of the models. Therefore, three backtesting procedures were used for the VaR, namely, the Kupiec’s test, a duration-based backtest, and an asymmetric VaR loss function. Similarly, three backtests were employed for the ES: a regression-based backtesting procedure, the Exceedance Residual backtest and the V-Tests. The validation results show that non-normal distributions are best suited for both model fitting and forecasting. The MC-GARCH(1,1) model under the Generalised Error Distribution (GED) innovation assumption gave the best fit to the intraday data and gave the best results for the ES forecasts. However, the asymmetric Skewed Student’s-t distribution for the innovation process provided the best results for the VaR forecasts. This paper presents the results of the first empirical study (to the best of the authors’ knowledge) in: (1) forecasting the intraday Expected Shortfall (ES) under different distributional assumptions for the MC-GARCH model; (2) assessing the MC-GARCH model under the Generalised Error Distribution (GED) innovation; (3) evaluating and ranking the VaR predictability of the MC-GARCH models using an asymmetric loss function.

Purpose This study aims to use a novel methodology to investigate the performance of several multivariate value at risk (VaR) and expected shortfall (ES) models implemented to assess the risk of an equally weighted portfolio consisting of high-frequency (1-min) observations for five foreign currencies, namely, EUR/USD, GBP/USD, EUR/JPY, USD/JPY and GBP/JPY. Design/methodology/approach By applying the multiplicative component generalised autoregressive conditional heteroskedasticity (MC-GARCH) model on each return series and by modelling the dependence structure using copulas, the 95 per cent intraday portfolio VaR and ES are forecasted for an out-of-sample set using Monte Carlo simulation. Findings In terms of VaR forecasting performance, the backtesting results indicated that four out of the five models implemented could not be rejected at 5 per cent level of significance. However, when the models were further evaluated for their ES forecasting power, only the Student’s t and Clayton models could not be rejected. The fact that some ES models were rejected at 5 per cent significance level highlights the importance of selecting an appropriate copula model for the dependence structure. Originality/value To the best of the authors’ knowledge, this is the first study to use the MC-GARCH and copula models to forecast, for the next 1 min, the VaR and ES of an equally weighted portfolio of foreign currencies. It is also the first study to analyse the performance of the MC-GARCH model under seven distributional assumptions for the innovation term.

We applied multiplicative GARCH-MIDAS two component models for price return volatility of selected commodities traded at the Ethiopian commodity exchange (ECX). Unlike the ‘traditional’ generalized autoregressive conditional heteroscedasticity (GARCH) family models, GARCH-MIDAS component model can capture the time-varying conditional as well as unconditional volatilities, and accommodates macroeconomic variables observed at different frequencies through mixed interval data sampling (MIDAS) specification. The results of our specification tests revealed the existence of both time-varying conditional and unconditional variance. The fitted GARCH-MIDAS component models showed that realized volatility, inflation rate and fuel oil price have had an increasing effect on the price volatility of the commodities under consideration, while real effective exchange rate (REER) had the opposite effect. Furthermore, mean square error (MSE), mean absolute error (MAE) and Diebold and Mariano (DM) test were used for evaluating and comparing the forecasting ability of GARCH-MIDAS component models against standard GARCH models. The results revealed that GARCH-MIDAS component models outperformed the standard GARCH model for high-frequency data.

This paper focuses on building statistical models to capture and forecast the traffic of mobile communication network in Vietnam. Following BoxJenkins method, a multiplicative seasonal ARIMA model is constructed to represent the mean component using the past values of traffic, a GARCH model is then incorporated to represent its volatility. The traffic is collected from EVN Telecom mobile communication network. The numerical result comparisons show that the multiplicative seasonal ARIMA/GARCH model built in this paper gives a better estimate when dealing with volatility clustering in the data series. However, in short-term prediction where the volatility has an insignificant influence, the achieved ARIMA model also can be considered as a good model to capture well the characteristics of EVN traffic series and gives reasonable forecasting results.

Nous nous intéressons à l'étude des propriétés théoriques des équations récurrentes stochastiques (SRE) et de leurs applications en finance. Ces modèles sont couramment utilisés en économétrie, y compris en économétrie de la finance, pour styliser la dynamique d'une variété de processus tels que la volatilité des rendements financiers. Cependant, la structure de probabilité ainsi que les propriétés statistiques de ces modèles sont encore mal connues, particulièrement lorsque le modèle est considéré en dimension infinie ou lorsqu'il est généré par un processus non indépendant. Ces deux caractéristiques entraînent de formidables difficultés à l'étude théorique de ces modèles. Dans ces contextes, nous nous intéressons à l'existence de solutions stationnaires, ainsi qu'aux propriétés statistiques et probabilistes de ces solutions.Nous établissons de nouvelles propriétés sur la trajectoire de la solution stationnaire des SREs que nous exploitons dans l'étude des propriétés asymptotiques de l'estimateur du quasi-maximum de vraisemblance (QMLE) des modèles de volatilité conditionnelle de type GARCH. En particulier, nous avons étudié la stationnarité et l'inférence statistique des modèles GARCH(p,q) semi-forts dans lesquels le processus d'innovation n'est pas nécessairement indépendant. Nous établissons la consistance du QMLE des GARCH (p,q) semi-forts sans hypothèses d'existence de moment, couramment supposée pour ces modèles, sur la distribution stationnaire. De même, nous nous sommes intéressés aux modèles GARCH à deux facteurs (GARCH-MIDAS); un facteur de volatilité à long terme et un autre à court terme. Ces récents modèles introduits par Engle et al. (2013) ont la particularité d'avoir des solutions stationnaires avec des distributions à queue épaisse. Ces modèles sont maintenant fréquemment utilisés en économétrie, cependant, leurs propriétés statistiques n'ont pas reçu beaucoup d'attention jusqu'à présent. Nous montrons la consistance et la normalité asymptotique du QMLE des modèles GARCH-MIDAS et nous proposons différentes procédures de test pour évaluer la présence de volatilité à long terme dans ces modèles. Nous illustrons nos résultats avec des simulations et des applications sur des données financières réelles.Enfin, nous étendons le résultat de Kesten (1975) sur le taux de croissance des séquences additives aux processus superadditifs. Nous déduisons de ce résultat des généralisations de la propriété de contraction des matrices aléatoires aux produits d'opérateurs stochastiques. Nous utilisons ces résultats pour établir des conditions nécessaires et suffisantes d'existence de solutions stationnaires du modèle affine à coefficients positifs des SREs dans l'espace des fonctions continues. Cette classe de modèles regroupe la plupart des modèles de volatilité conditionnelle, y compris les GARCH fonctionnels
We are interested in the theoretical properties of Stochastic Recurrent Equations (SRE) and their applications in finance. These models are widely used in econometrics, including financial econometrics, to explain the dynamics of various processes such as the volatility of financial returns. However, the probability structure and statistical properties of these models are still not well understood, especially when the model is considered in infinite dimensions or driven by non-independent processes. These two features lead to significant difficulties in the theoretical study of these models. In this context, we aim to explore the existence of stationary solutions and the statistical and probabilistic properties of these solutions.We establish new properties on the trajectory of the stationary solution of SREs, which we use to study the asymptotic properties of the quasi-maximum likelihood estimator (QMLE) of GARCH-type (generalized autoregressive conditional heteroskedasticity) conditional volatility models. In particular, we study the stationarity and statistical inference of semi-strong GARCH(p,q) models where the innovation process is not necessarily independent. We establish the consistency of the QMLE of semi-strong GARCHs without assuming the commonly used condition that the stationary distribution admits a small-order moment. In addition, we are interested in the two-factor volatility GARCH models (GARCH-MIDAS); a long-run, and a short-run volatility. These models were recently introduced by Engle et al. (2013) and have the particularity to admit stationary solutions with heavy-tailed distributions. These models are now widely used but their statistical properties have not received much attention. We show the consistency and asymptotic normality of the QMLE of the GARCH-MIDAS models and provide various test procedures to evaluate the presence of long-run volatility in these models. We also illustrate our results with simulations and applications to real financial data.Finally, we extend a result of Kesten (1975) on the growth rate of additive sequences to superadditive processes. From this result, we derive generalizations of the contraction property of random matrices to products of stochastic operators. We use these results to establish necessary and sufficient conditions for the existence of stationary solutions of the affine case with positive coefficients of SREs in the space of continuous functions. This class of models includes most conditional volatility models, including functional GARCHs

The research presented in this thesis addresses different aspects of dynamic portfolio construction and portfolio risk measurement. It brings the research on dynamic portfolio optimization, replicating portfolio construction, dynamic portfolio risk measurement and volatility forecast together. The overall aim of this research is threefold. First, it is aimed to examine the portfolio construction and risk measurement performance of a broad set of volatility forecast and portfolio optimization model. Second, in an effort to improve their forecast accuracy and portfolio construction performance, it is aimed to propose new models or new formulations to the available models. Third, in order to enhance the replication performance of hedge fund returns, it is aimed to introduce a replication approach that has the potential to be used in numerous applications, in investment management. In order to achieve these aims, Chapter 2 addresses risk measurement in dynamic portfolio construction. In this chapter, further evidence on the use of multivariate conditional volatility models in hedge fund risk measurement and portfolio allocation is provided by using monthly returns of hedge fund strategy indices for the period 1990 to 2009. Building on Giamouridis and Vrontos (2007), a broad set of multivariate GARCH models, as well as, the simpler exponentially weighted moving average (EWMA) estimator of RiskMetrics (1996) are considered. It is found that, while multivariate GARCH models provide some improvements in portfolio performance over static models, they are generally dominated by the EWMA model. In particular, in addition to providing a better risk-adjusted performance, the EWMA model leads to dynamic allocation strategies that have a substantially lower turnover and could therefore be expected to involve lower transaction costs. Moreover, it is shown that these results are robust across the low - volatility and high-volatility sub-periods. Chapter 3 addresses optimization in dynamic portfolio construction. In this chapter, the advantages of introducing alternative optimization frameworks over the mean-variance framework in constructing hedge fund portfolios for a fund of funds. Using monthly return data of hedge fund strategy indices for the period 1990 to 2011, the standard mean-variance approach is compared with approaches based on CVaR, CDaR and Omega, for both conservative and aggressive hedge fund investors. In order to estimate portfolio CVaR, CDaR and Omega, a semi-parametric approach is proposed, in which first the marginal density of each hedge fund index is modelled using extreme value theory and the joint density of hedge fund index returns is constructed using a copula-based approach. Then hedge fund returns from this joint density are simulated in order to compute CVaR, CDaR and Omega. The semi-parametric approach is compared with the standard, non-parametric approach, in which the quantiles of the marginal density of portfolio returns are estimated empirically and used to compute CVaR, CDaR and Omega. Two main findings are reported. The first is that CVaR-, CDaR- and Omega-based optimization offers a significant improvement in terms of risk-adjusted portfolio performance over mean-variance optimization. The second is that, for all three risk measures, semi-parametric estimation of the optimal portfolio offers a very significant improvement over non-parametric estimation. The results are robust to as the choice of target return and the estimation period. Chapter 4 searches for improvements in portfolio risk measurement by addressing volatility forecast. In this chapter, two new univariate Markov regime switching models based on intraday range are introduced. A regime switching conditional volatility model is combined with a robust measure of volatility based on intraday range, in a framework for volatility forecasting. This chapter proposes a one-factor and a two-factor model that combine useful properties of range, regime switching, nonlinear filtration, and GARCH frameworks. Any incremental improvement in the performance of volatility forecasting is searched for by employing regime switching in a conditional volatility setting with enhanced information content on true volatility. Weekly S&P500 index data for 1982-2010 is used. Models are evaluated by using a number of volatility proxies, which approximate true integrated volatility. Forecast performance of the proposed models is compared to renowned return-based and range-based models, namely EWMA of Riskmetrics, hybrid EWMA of Harris and Yilmaz (2009), GARCH of Bollerslev (1988), CARR of Chou (2005), FIGARCH of Baillie et al. (1996) and MRSGARCH of Klaassen (2002). It is found that the proposed models produce more accurate out of sample forecasts, contain more information about true volatility and exhibit similar or better performance when used for value at risk comparison. Chapter 5 searches for improvements in risk measurement for a better dynamic portfolio construction. This chapter proposes multivariate versions of one and two factor MRSACR models introduced in the fourth chapter. In these models, useful properties of regime switching models, nonlinear filtration and range-based estimator are combined with a multivariate setting, based on static and dynamic correlation estimates. In comparing the out-of-sample forecast performance of these models, eminent return and range-based volatility models are employed as benchmark models. A hedge fund portfolio construction is conducted in order to investigate the out-of-sample portfolio performance of the proposed models. Also, the out-of-sample performance of each model is tested by using a number of statistical tests. In particular, a broad range of statistical tests and loss functions are utilized in evaluating the forecast performance of the variance covariance matrix of each portfolio. It is found that, in terms statistical test results, proposed models offer significant improvements in forecasting true volatility process, and, in terms of risk and return criteria employed, proposed models perform better than benchmark models. Proposed models construct hedge fund portfolios with higher risk-adjusted returns, lower tail risks, offer superior risk-return tradeoffs and better active management ratios. However, in most cases these improvements come at the expense of higher portfolio turnover and rebalancing expenses. Chapter 6 addresses the dynamic portfolio construction for a better hedge fund return replication and proposes a new approach. In this chapter, a method for hedge fund replication is proposed that uses a factor-based model supplemented with a series of risk and return constraints that implicitly target all the moments of the hedge fund return distribution. The approach is used to replicate the monthly returns of ten broad hedge fund strategy indices, using long-only positions in ten equity, bond, foreign exchange, and commodity indices, all of which can be traded using liquid, investible instruments such as futures, options and exchange traded funds. In out-of-sample tests, proposed approach provides an improvement over the pure factor-based model, offering a closer match to both the return performance and risk characteristics of the hedge fund strategy indices.