Large Volatility Matrix Research Articles

Factor Overnight GARCH-Itô Models

Abstract This article introduces a unified factor overnight GARCH-Itô model for large volatility matrix estimation and prediction. To account for whole-day market dynamics, the proposed model has two different instantaneous factor volatility processes for the open-to-close and close-to-open periods, while each embeds the discrete-time multivariate GARCH model structure. To estimate latent factor volatility, we assume the low rank plus sparse structure and employ nonparametric estimation procedures. Then, based on the connection between the discrete-time model structure and the continuous-time diffusion process, we propose a weighted least squares estimation procedure with the non-parametric factor volatility estimator and establish its asymptotic theorems.

Adaptive robust large volatility matrix estimation based on high-frequency financial data

Several novel statistical methods have been developed to estimate large integrated volatility matrices based on high-frequency financial data. To investigate their asymptotic behaviors, they require a sub-Gaussian or finite high-order moment assumption for observed log-returns, which cannot account for the heavy-tail phenomenon of stock-returns. Recently, a robust estimator was developed to handle heavy-tailed distributions with some bounded fourth-moment assumption. However, we often observe that log-returns have heavier tail distribution than the finite fourth-moment and that the degrees of heaviness of tails are heterogeneous across asset and over time. In this paper, to deal with the heterogeneous heavy-tailed distributions, we develop an adaptive robust integrated volatility estimator that employs pre-averaging and truncation schemes based on jump-diffusion processes. We call this an adaptive robust pre-averaging realized volatility (ARP) estimator. We show that the ARP estimator has a sub-Weibull tail concentration with only finite 2α-th moments for any α>1. In addition, we establish matching upper and lower bounds to show that the ARP estimation procedure is optimal. To estimate large integrated volatility matrices using the approximate factor model, the ARP estimator is further regularized using the principal orthogonal complement thresholding (POET) method. The numerical study is conducted to check the finite sample performance of the ARP estimator.

Large volatility matrix analysis using global and national factor models

Several large volatility matrix inference procedures have been developed, based on the latent factor model. They often assumed that there are a few of common factors, which can account for volatility dynamics. However, several studies have demonstrated the presence of local factors. In particular, when analyzing the global stock market, we often observe that nation-specific factors explain their own country’s volatility dynamics. To account for this, we propose the Double Principal Orthogonal complEment Thresholding (Double-POET) method, based on multi-level factor models, and also establish its asymptotic properties. Furthermore, we demonstrate the drawback of using the regular principal orthogonal component thresholding (POET) when the local factor structure exists. We also describe the blessing of dimensionality using Double-POET for local covariance matrix estimation. Finally, we investigate the performance of the Double-POET estimator in an out-of-sample portfolio allocation study using international stocks from 20 financial markets.

Multivariate Overnight GARCH-Ito Model with Applications in Large Volatility Matrix Estimation and Prediction

This paper introduces a unified multivariate overnight GARCH-It\^o model for volatility matrix estimation and prediction both in the low- and high-dimensional set-up. To account for whole-day market dynamics in the financial market, the proposed model has two different instantaneous volatility processes for the open-to-close and close-to-open periods, while each embeds the discrete-time multivariate GARCH model structure. We call it the multivariate overnight GARCH-It\^o (MOGI) model. Based on the connection between the discrete-time model structure and the continuous-time diffusion process, we propose a weighted least squares estimation procedure for estimating model parameters with open-to-close high-frequency and close-to-open low-frequency financial data, and establish its asymptotic theorems. We further discuss the prediction of future vast volatility matrices and study its asymptotic properties. A simulation study is conducted to check the finite sample performance of the proposed estimation and prediction methods. The empirical analysis is carried out to compare the performance of the proposed MOGI model with other benchmark methods in portfolio allocation problems.

Factor and Idiosyncratic VAR-Ito Volatility Models for Heavy-Tailed High-Frequency Financial Data

Various parametric models have been developed to predict large volatility matrices, based on the approximate factor model structure. They mainly focus on the dynamics of the factor volatility with some finite high-order moment assumptions. However, the empirical studies have shown that the idiosyncratic volatility also has a dynamic structure and it comprises a large proportion of the total volatility. Furthermore, we often observe that the financial market exhibits heavy tails. To account for these stylized features in financial returns, we introduce a novel Ito diffusion process for both factor and idiosyncratic volatilities whose eigenvalues follow the vector auto-regressive (VAR) model. We call it the factor and idiosyncratic VAR-Ito (FIVAR-Ito) model. To handle the heavy-tailedness and curse of dimensionality, we propose a robust parameter estimation method for a high-dimensional VAR model. We apply the robust estimator to predicting large volatility matrices and investigate its asymptotic properties. Simulation studies are conducted to validate the finite sample performance of the proposed estimation and prediction methods. Using high-frequency trading data, we apply the proposed method to large volatility matrix prediction and minimum variance portfolio allocation and showcase the new model and the proposed method.

Structured volatility matrix estimation for non-synchronized high-frequency financial data

Several large volatility matrix estimation procedures have been recently developed for factor-based Itô processes whose integrated volatility matrix consists of low-rank and sparse matrices. Their performance depends on the accuracy of input volatility matrix estimators. When estimating co-volatilities based on high-frequency data, one of the crucial challenges is non-synchronization for illiquid assets, which makes their co-volatility estimators inaccurate. In this paper, we study how to estimate the large integrated volatility matrix without using co-volatilities of illiquid assets. Specifically, we pretend that the co-volatilities for illiquid assets are missing, and estimate the low-rank matrix using a matrix completion scheme with a structured missing pattern. To further regularize the sparse volatility matrix, we employ the principal orthogonal complement thresholding method (POET). We also investigate the asymptotic properties of the proposed estimation procedure and demonstrate its advantages over using co-volatilities of illiquid assets. The advantages of our methods are also verified by an extensive simulation study and illustrated by high-frequency data for NYSE stocks.

Factor GARCH-Itô models for high-frequency data with application to large volatility matrix prediction

Several novel large volatility matrix estimation methods have been developed based on the high-frequency financial data. They often employ the approximate factor model that leads to a low-rank plus sparse structure for the integrated volatility matrix and facilitates estimation of large volatility matrices. However, for predicting future volatility matrices, these nonparametric estimators do not have a dynamic structure to implement. In this paper, we introduce a novel Itô diffusion process based on the approximate factor models and call it a factor GARCH-Itô model. We then investigate its properties and propose a quasi-maximum likelihood estimation method for the parameter of the factor GARCH-Itô model. We also apply it to estimating conditional expected large volatility matrices and establish their asymptotic properties. Simulation studies are conducted to validate the finite sample performance of the proposed estimation methods. The proposed method is also illustrated by using data from the constituents of the S&P 500 index and an application to constructing the minimum variance portfolio with gross exposure constraints.

Large volatility matrix estimation with factor-based diffusion model for high-frequency financial data

Large volatility matrices are involved in many finance practices, and estimating large volatility matrices based on high-frequency financial data encounters the “curse of dimensionality”. It is a common approach to impose a sparsity assumption on the large volatility matrices to produce consistent volatility matrix estimators. However, due to the existence of common factors, assets are highly correlated with each other, and it is not reasonable to assume the volatility matrices are sparse in financial applications. This paper incorporates factor influence in the asset pricing model and investigates large volatility matrix estimation under the factor price model together with some sparsity assumption. We propose to model asset prices by assuming that asset prices are governed by common factors and that the assets with similar characteristics share the same association with the factors. We then impose some reasonable sparsity condition on the part of the volatility matrices after accounting for the factor contribution. Under the proposed factor-based model and sparsity assumption, we develop an estimation scheme called “blocking and regularizing”. Asymptotic properties of the proposed estimator are studied, and its finite sample performance is tested via extensive numerical studies to support theoretical results.

Robust High-Dimensional Volatility Matrix Estimation for High-Frequency Factor Model

ABSTRACTHigh-frequency financial data allow us to estimate large volatility matrices with relatively short time horizon. Many novel statistical methods have been introduced to address large volatility matrix estimation problems from a high-dimensional Itô process with microstructural noise contamination. Their asymptotic theories require sub-Gaussian or some finite high-order moments assumptions for observed log-returns. These assumptions are at odd with the heavy tail phenomenon that is pandemic in financial stock returns and new procedures are needed to mitigate the influence of heavy tails. In this article, we introduce the Huber loss function with a diverging threshold to develop a robust realized volatility estimation. We show that it has the sub-Gaussian concentration around the volatility with only finite fourth moments of observed log-returns. With the proposed robust estimator as input, we further regularize it by using the principal orthogonal component thresholding (POET) procedure to estimate the large volatility matrix that admits an approximate factor structure. We establish the asymptotic theories for such low-rank plus sparse matrices. The simulation study is conducted to check the finite sample performance of the proposed estimation methods.

On a spiked model for large volatility matrix estimation from noisy high-frequency data

Recently, inference about high-dimensional integrated covariance matrices (ICVs) based on noisy high-frequency data has emerged as a challenging problem. In the literature, a pre-averaging estimator (PA-RCov) is proposed to deal with the microstructure noise. Using the large-dimensional random matrix theory, it has been established that the eigenvalue distribution of the PA-RCov matrix is intimately linked to that of the ICV through the Marčenko–Pasturequation. Consequently, the spectrum of the ICV can be inferred from that of the PA-RCov. However, extensive data analyses demonstrate that the spectrum of the PA-RCov is spiked, that is, a few large eigenvalues (spikes) stay away from the others which form a rather continuous distribution with a density function (bulk). Therefore, any inference on the ICVs must take into account this spiked structure. As a methodological contribution, a spiked model is proposed for the ICVs where spikes can be inferred from those of the available PA-RCov matrices. The consistency of the inference procedure is established. In addition, the methodology is applied to the real data from the US and Hong Kong markets. It is found that the model clearly outperforms the existing one in predicting the existence of spikes and in mimicking the empirical PA-RCov matrices.

Factor GARCH-ITO Models for High-Frequency Data with Application to Large Volatility Matrix Prediction

Several novel large volatility matrix estimation methods have been developed based on the high-frequency financial data. They often employ the approximate factor model that leads to a low-rank plus sparse structure for the integrated volatility matrix and facilitates estimation of large volatility matrices. However, for predicting future volatility matrices, these nonparametric estimators do not have a dynamic structure to implement. In this paper, we introduce a novel Ito diffusion process based on the approximate factor models and call it a factor GARCH-Ito model. We then investigate its properties and propose a quasi-maximum likelihood estimation method for the parameter of the factor GARCH-Ito model. We also apply it to estimating conditional expected large volatility matrices and establish their asymptotic properties. Simulation studies are conducted to validate the finite sample performance of the proposed estimation methods. The proposed method is also illustrated by using data from the constituents of the S&P 500 index and an application to constructing the minimum variance portfolio with gross exposure constraints.

Robust High-Dimensional Volatility Matrix Estimation for High-Frequency Factor Model

High-frequency financial data allow us to estimate large volatility matrices with relatively short time horizon. Many novel statistical methods have been introduced to address large volatility matrix estimation problems from a high-dimensional Ito process with microstructural noise contamination. Their asymptotic theories require sub-Gaussian or some finite high-order moments assumptions for observed log-returns. These assumptions are at odd with the heavy tail phenomenon that is pandemic in financial stock returns and new procedures are needed to mitigate the influence of heavy tails. In this paper, we introduce the Huber loss function with a diverging threshold to develop a robust realized volatility estimation. We show that it has the sub-Gaussian concentration around the volatility with only finite fourth moments of observed log-returns. With the proposed robust estimator as input, we further regularize it by using the principal orthogonal component thresholding (POET) procedure to estimate the large volatility matrix that admits an approximate factor structure. We establish the asymptotic theories for such low-rank plus sparse matrices. The simulation study is conducted to check the finite sample performance of the proposed estimation methods.

Structured Volatility Matrix Estimation for Non-Synchronized High-Frequency Financial Data

Recently several large volatility matrix estimation procedures have been developed for factor-based Ito processes whose integrated volatility matrix consists of low-rank and sparse matrices. Their performance depends on the accuracy of input volatility matrix estimators. When estimating co-volatilities based on high-frequency data, one of the crucial challenges is non-synchronization for illiquid assets, which makes their co-volatility estimators inaccurate. In this paper, we study how to estimate the large integrated volatility matrix without using co-volatilities of illiquid assets. Specifically, we pretend that the co-volatilities for illiquid assets are missing, and estimate the low-rank matrix using a matrix completion scheme with a structured missing pattern. To further regularize the sparse volatility matrix, we employ the principal orthogonal complement thresholding method (POET). We also investigate the asymptotic properties of the proposed estimation procedure and demonstrate its advantages over using co-volatilities of illiquid assets. The advantages of our methods are also verified by an extensive simulation study and illustrated by high-frequency financial data for constituents of the S&P 500 index.

Adaptive thresholding for large volatility matrix estimation based on high-frequency financial data

Universal thresholding methods have been developed to estimate the large sparse integrated volatility matrix of underlying assets based on high-frequency financial data. Since the integrated volatility matrix often has entries with a wide range of variability, universal thresholding estimators do not take the varying entries into consideration and may have unsatisfactory performances. This paper investigates adaptive thresholding estimation of the large integrated volatility matrix. We first construct an estimator for the asymptotic variance of the pre-averaging realized volatility estimator and then use the two estimators to develop an adaptive thresholding estimator of the large volatility matrix. It is shown that the adaptive thresholding estimator can achieve the optimal rate of convergence over the class of the sparse integrated volatility matrix when both the number of assets and sample size are allowed to go to infinity, while the universal thresholding estimator can achieve only the sub-optimal convergence rate. Also we discuss how to harness the adaptive thresholding scheme in the approximate factor model. The simulation study is conducted to check the finite sample performance of the adaptive thresholding estimators.

Asymptotic theory for large volatility matrix estimation based on high-frequency financial data

In financial practices and research studies, we often encounter a large number of assets. The availability of high-frequency financial data makes it possible to estimate the large volatility matrix of these assets. Existing volatility matrix estimators such as kernel realized volatility and pre-averaging realized volatility perform poorly when the number of assets is very large, and in fact they are inconsistent when the number of assets and sample size go to infinity. In this paper, we introduce threshold rules to regularize kernel realized volatility, pre-averaging realized volatility, and multi-scale realized volatility. We establish asymptotic theory for these threshold estimators in the framework that allows the number of assets and sample size to go to infinity. Their convergence rates are derived under sparsity on the large integrated volatility matrix. In particular we have shown that the threshold kernel realized volatility and threshold pre-averaging realized volatility can achieve the optimal rate with respect to the sample size through proper bias corrections, but the bias adjustments cause the estimators to lose positive semi-definiteness; on the other hand, in order to be positive semi-definite, the threshold kernel realized volatility and threshold pre-averaging realized volatility have slower convergence rates with respect to the sample size. A simulation study is conducted to check the finite sample performances of the proposed threshold estimators with over hundred assets.

Volatility analysis in high‐frequency financial data

In this paper we provide a detailed review of univariate and multivariate volatility analysis within the framework of high‐frequency financial data setting. The field of volatility modeling and analysis for high‐frequency financial data has experienced a rapid development over the past decade. High‐frequency financial data pose tremendous challenges on volatility modeling and analysis, such as microstructure noise, non‐synchronization, irregularly spaced times between observations, nonstationarity, and jumps. We present models for asset prices and high‐frequency financial data. We discuss common volatility estimators such as realized volatility, two‐time scale and multi‐scale realized volatility estimators, realized kernel volatility estimator and pre‐averaging volatility estimator in the univariate setting. For estimating co‐volatility, we illustrate data synchronization methods including previous tick, refresh time, and generalized sampling time and present two‐time scale and multi‐scale realized volatility matrix estimators, realized kernel volatility estimator, and the quasi‐maximum likelihood method built on the synchronized data. We also feature pre‐averaging volatility and Hayashi‐Yoshida estimators which are based on inexplicit data synchronization. We provide detailed formulation and properties for each estimator or method and point out their relationships and differences. Given the recent interests in big data and high‐dimensional statistics, large volatility matrix inference approaches are highlighted in the end. WIREs Comput Stat 2014, 6:393–404. doi: 10.1002/wics.1330For further resources related to this article, please visit the WIREs website.This article is categorized under: Applications of Computational Statistics > Computational Finance Statistical and Graphical Methods of Data Analysis > Nonparametric Methods Data: Types and Structure > Time Series, Stochastic Processes, and Functional Data

FAST CONVERGENCE RATES IN ESTIMATING LARGE VOLATILITY MATRICES USING HIGH-FREQUENCY FINANCIAL DATA

Financial practices often need to estimate an integrated volatility matrix of a large number of assets using noisy high-frequency data. Many existing estimators of a volatility matrix of small dimensions become inconsistent when the size of the matrix is close to or larger than the sample size. This paper introduces a new type of large volatility matrix estimator based on nonsynchronized high-frequency data, allowing for the presence of microstructure noise. When both the number of assets and the sample size go to infinity, we show that our new estimator is consistent and achieves a fast convergence rate, where the rate is optimal with respect to the sample size. A simulation study is conducted to check the finite sample performance of the proposed estimator.

Large Volatility Matrix Inference via Combining Low-Frequency and High-Frequency Approaches

It is increasingly important in financial economics to estimate volatilities of asset returns. However, most of the available methods are not directly applicable when the number of assets involved is large, due to the lack of accuracy in estimating high-dimensional matrices. Therefore it is pertinent to reduce the effective size of volatility matrices in order to produce adequate estimates and forecasts. Furthermore, since high-frequency financial data for different assets are typically not recorded at the same time points, conventional dimension-reduction techniques are not directly applicable. To overcome those difficulties we explore a novel approach that combines high-frequency volatility matrix estimation together with low-frequency dynamic models. The proposed methodology consists of three steps: (i) estimate daily realized covolatility matrices directly based on high-frequency data, (ii) fit a matrix factor model to the estimated daily covolatility matrices, and (iii) fit a vector autoregressive model to the estimated volatility factors. We establish the asymptotic theory for the proposed methodology in the framework that allows sample size, number of assets, and number of days go to infinity together. Our theory shows that the relevant eigenvalues and eigenvectors can be consistently estimated. We illustrate the methodology with the high-frequency price data on several hundreds of stocks traded in Shenzhen and Shanghai Stock Exchanges over a period of 177 days in 2003. Our approach pools together the strengths of modeling and estimation at both intra-daily (high-frequency) and inter-daily (low-frequency) levels.