Forecasting volatility and co-volatility of crude oil and gold futures: Effects of leverage,jumps, spillovers,and geopolitical risks

To forecast the covariance matrix for the returns of crude oil and gold futures, this paper examines the effects of leverage, jumps, spillovers, and geopolitical risks by using their respective realized covariance matrices. To guarantee the positive definiteness of the forecasts, we consider the full BEKK structure on the conditional Wishart model. By the specification, we can flexibly divide the direct and spillover effects of volatility feedback, negative returns, and jumps. The empirical analysis indicates the benefits of accommodating the spillover effects of negative returns, and the geopolitical risks indicator for modeling and forecasting the covariance matrix.     

Bayesian parametric and semiparametric factor models for large realized covariance matrices

This paper introduces a new factor structure suitable for modeling large realized covariance matrices with full likelihood‐based estimation. Parametric and nonparametric versions are introduced. Because of the computational advantages of our approach, we can model the factor nonparametrically as a Dirichlet process mixture or as an infinite hidden Markov mixture, which leads to an infinite mixture of inverse‐Wishart distributions. Applications to 10 assets and 60 assets show that the models perform well. By exploiting parallel computing the models can be estimated in a matter of a few minutes.     

Conditions for invariance of the multivariate versions of grubbs’ test and bartlett’s test under a general dependency structure

Summary This paper generalizes a result by Stadje (1984) by deriving conditions for which a general dependency structure for multivariate observations, given in Pavur (1987), yields a positive definite covariance structure. This general dependency structure allows the sample covariance matrix to be distributed as a constant times a Wishart random matrix. It is then demonstrated that the maximum squared-radii test and a test for equal population covariance matrices have null distributions which remain unchanged when the new general dependency structure, rather than the usual independence structure, for the vector observations, is assumed. Moreover, under a general dependency structure for which the population covariance matrices are unequal, it is shown that the distribution of the test statistic for testing equal covariance matrices is identical to the distribution of the same test statistic when the population covariance matrices are equal and the observations are independent.     

The Wishart Autoregressive process of multivariate stochastic volatility

The Wishart Autoregressive (WAR) process is a dynamic model for time series of multivariate stochastic volatility. The WAR naturally accommodates the positivity and symmetry of volatility matrices and provides closed-form non-linear forecasts. The estimation of the WAR is straighforward, as it relies on standard methods such as the Method of Moments and Maximum Likelihood. For illustration, the WAR is applied to a sequence of intraday realized volatility–covolatility matrices from the Toronto Stock Market (TSX).     

Bootstrapping realized multivariate volatility measures

We propose a bootstrap method for statistics that are a function of multivariate high frequency returns such as realized regression, covariance and correlation coefficients. We show that the finite sample performance of the bootstrap is superior to the existing first-order asymptotic theory. Nevertheless, and contrary to the existing results in the bootstrap literature for regression models subject to error heteroskedasticity, the Edgeworth expansion for the pairs bootstrap that we develop here shows that this method is not second-order accurate. We argue that this is due to the fact that the conditional mean parameters of realized regression models are heterogeneous under stochastic volatility.     

Modeling and Forecasting Large Realized Covariance Matrices and Portfolio Choice

We consider modeling and forecasting large realized covariance matrices by penalized vector autoregressive models. We consider Lasso‐type estimators to reduce the dimensionality and provide strong theoretical guarantees on the forecast capability of our procedure. We show that we can forecast realized covariance matrices almost as precisely as if we had known the true driving dynamics of these in advance. We next investigate the sources of these driving dynamics as well as the performance of the proposed models for forecasting the realized covariance matrices of the 30 Dow Jones stocks. We find that the dynamics are not stable as the data are aggregated from the daily to lower frequencies. Furthermore, we are able beat our benchmark by a wide margin. Finally, we investigate the economic value of our forecasts in a portfolio selection exercise and find that in certain cases an investor is willing to pay a considerable amount in order get access to our forecasts. Copyright © 2016 John Wiley & Sons, Ltd.     

Forecasting multivariate realized stock market volatility

We present a new matrix-logarithm model of the realized covariance matrix of stock returns. The model uses latent factors which are functions of lagged volatility, lagged returns and other forecasting variables. The model has several advantages: it is parsimonious; it does not require imposing parameter restrictions; and, it results in a positive-definite estimated covariance matrix. We apply the model to the covariance matrix of size-sorted stock returns and find that two factors are sufficient to capture most of the dynamics.     

Dynamic correlation multivariate stochastic volatility with latent factors

Modeling the correlation structure of returns is essential in many financial applications. Considerable evidence from empirical studies has shown that the correlation among asset returns is not stable over time. A recent development in the multivariate stochastic volatility literature is the application of inverse Wishart processes to characterize the evolution of return correlation matrices. Within the inverse Wishart multivariate stochastic volatility framework, we propose a flexible correlated latent factor model to achieve dimension reduction and capture the stylized fact of ‘correlation breakdown’ simultaneously. The parameter estimation is based on existing Markov chain Monte Carlo methods. We illustrate the proposed model with several empirical studies. In particular, we use high‐dimensional stock return data to compare our model with competing models based on multiple performance metrics and tests. The results show that the proposed model not only describes historic stylized facts reasonably but also provides the best overall performance.     

Missing in Asynchronicity: A Kalman‐em Approach for Multivariate Realized Covariance Estimation

Motivated by the need for a positive‐semidefinite estimator of multivariate realized covariance matrices, we model noisy and asynchronous ultra‐high‐frequency asset prices in a state‐space framework with missing data. We then estimate the covariance matrix of the latent states through a Kalman smoother and expectation maximization (KEM) algorithm. Iterating between the two EM steps, we obtain a covariance matrix estimate which is robust to both asynchronicity and microstructure noise, and positive‐semidefinite by construction. We show the performance of the KEM estimator using extensive Monte Carlo simulations that mimic the liquidity and market microstructure characteristics of the S&P 500 universe as well as in a high‐dimensional application on US stocks. KEM provides very accurate covariance matrix estimates and significantly outperforms alternative approaches recently introduced in the literature. Copyright © 2014 John Wiley & Sons, Ltd.     

Markovian lifts of positive semidefinite affine Volterra-type processes

We consider stochastic partial differential equations appearing as Markovian lifts of matrix-valued (affine) Volterra-type processes from the point of view of the generalized Feller property (see, e.g., Dörsek and Teichmann in A semigroup point of view on splitting schemes for stochastic (partial) differential equations, 2010. arXiv:1011.2651). We introduce in particular Volterra Wishart processes with fractional kernels and values in the cone of positive semidefinite matrices. They are constructed from matrix products of infinite dimensional Ornstein–Uhlenbeck processes whose state space is the set of matrix-valued measures. Parallel to that we also consider positive definite Volterra pure jump processes, giving rise to multivariate Hawkes-type processes. We apply these affine covariance processes for multivariate (rough) volatility modeling and introduce a (rough) multivariate Volterra Heston-type model.

     

More on connections between Wishart and matrix GIG distributions

The paper is devoted to relations between the matrix GIG and Wishart distributions. Our basic tool in the first part is a version of the Matsumoto-Yor property for matrix variables. This approach covers the following issues: the Herz identity for the Bessel function of matrix variate argument, characterization of a class of Wishart matrices and linear transformations of the matrix GIG distribution. The Bayesian Wishart model, studied in the second part, gives an alternative definition of the matrix GIG distribution. Such a model is characterized by linearity of conditional expectations and matrix GIG conditional distribution. It is also extended to Bayesian matrix GIG models, in the framework of which an interesting independence property is proved.     

Distribution-free tests of mean vectors and covariance matrices for multivariate paired data

We study a permutation procedure to test the equality of mean vectors, homogeneity of covariance matrices, or simultaneous equality of both mean vectors and covariance matrices in multivariate paired data. We propose to use two test statistics for the equality of mean vectors and the homogeneity of covariance matrices, respectively, and combine them to test the simultaneous equality of both mean vectors and covariance matrices. Since the combined test has composite null hypothesis, we control its type I error probability and theoretically prove the asymptotic unbiasedness and consistency of the combined test. The new procedure requires no structural assumption on the covariances. No distributional assumption is imposed on the data, except that the permutation test for mean vector equality assumes symmetric joint distribution of the paired data. We illustrate the good performance of the proposed approach with comparison to competing methods via simulations. We apply the proposed method to testing the symmetry of tooth size in a dental study and to finding differentially expressed gene sets with dependent structures in a microarray study of prostate cancer.     

Realized volatility forecast with the Bayesian random compressed multivariate HAR model

We develop a Bayesian random compressed multivariate heterogeneous autoregressive (BRC-MHAR) model to forecast the realized covariance matrices of stock returns. The proposed model randomly compresses the predictors and reduces the number of parameters. We also construct several competing multivariate volatility models with the alternative shrinkage methods to compress the parameter’s dimensions. We compare the forecast performances of the proposed models with the competing models based on both statistical and economic evaluations. The results of statistical evaluation suggest that the BRC-MHAR models have the better forecast precision than the competing models for the short-term horizon. The results of economic evaluation suggest that the BRC-MHAR models are superior to the competing models in terms of the average return, the Shape ratio and the economic value.     

Reinforced urn processes for credit risk models

We propose a Bayesian nonparametric model to estimate rating migration matrices and default probabilities using the reinforced urn processes (RUP) introduced in Muliere et al. (2000). The estimated default probability becomes our prior information in a parametric model for the prediction of the number of bankruptcies, with the only assumption of exchangeability within rating classes. The Polya urn construction of the transition matrix justifies a Beta distributed de Finetti measure. Dependence among the processes is introduced through the dependence among the default probabilities, with the Bivariate Beta Distribution proposed in Olkin and Liu (2003) and its multivariate generalization.     

Modelling and forecasting multivariate realized volatility

This paper proposes a methodology for dynamic modelling and forecasting of realized covariance matrices based on fractionally integrated processes. The approach allows for flexible dependence patterns and automatically guarantees positive definiteness of the forecast. We provide an empirical application of the model, which shows that it outperforms other approaches in the extant literature, both in terms of statistical precision as well as in terms of providing a superior mean‐variance trade‐off in a classical investment decision setting. Copyright © 2010 John Wiley & Sons, Ltd.     

Multivariate rotated ARCH models

This paper introduces a new class of multivariate volatility models which is easy to estimate using covariance targeting, even with rich dynamics. We call them rotated ARCH (RARCH) models. The basic structure is to rotate the returns and then to fit them using a BEKK-type parameterization of the time-varying covariance whose long-run covariance is the identity matrix. This yields the rotated BEKK (RBEKK) model. The extension to DCC-type parameterizations is given, introducing the rotated DCC (RDCC) model. Inference for these models is computationally attractive, and the asymptotics are standard. The techniques are illustrated using data on the DJIA stocks.     

Method of moments estimation of GO-GARCH models

We propose a new estimation method for the factor loading matrix in generalized orthogonal GARCH (GO-GARCH) models. The method is based on eigenvectors of suitably defined sample autocorrelation matrices of squares and cross-products of returns. The method is numerically more attractive than likelihood-based estimation. Furthermore, the new method does not require strict assumptions on the volatility models of the factors, and therefore is less sensitive to model misspecification. We provide conditions for consistency of the estimator, and study its efficiency relative to maximum likelihood estimation using Monte Carlo simulations. The method is applied to European sector returns.     

A Model Confidence Set approach to the combination of multivariate volatility forecasts

In predicting conditional covariance matrices of financial portfolios, practitioners are required to choose among several alternative options, facing a number of different sources of uncertainty. A first source is related to the frequency at which prices are observed, either daily or intradaily. Using prices sampled at higher frequency inevitably poses additional sources of uncertainty related to the selection of the optimal intradaily sampling frequency and to the construction of the best realized estimator. Likewise, the choices of model structure and estimation method also have a critical role. In order to alleviate the impact of these sources of uncertainty, we propose a forecast combination strategy based on the Model Confidence Set [MCS] to adaptively identify the set of most accurate predictors. The combined predictor is shown to achieve superior performance with respect to the whole model universe plus three additional competitors, independently of the MCS or portfolio settings.     

HAC estimation in a spatial framework

We suggest a non-parametric heteroscedasticity and autocorrelation consistent (HAC) estimator of the variance–covariance (VC) matrix for a vector of sample moments within a spatial context. We demonstrate consistency under a set of assumptions that should be satisfied by a wide class of spatial models. We allow for more than one measure of distance, each of which may be measured with error. Monte Carlo results suggest that our estimator is reasonable in finite samples. We then consider a spatial model containing various complexities and demonstrate that our HAC estimator can be applied in the context of that model.