Forecasting the weekly time-varying beta of UK firms: GARCH models vs. Kalman filter method

This paper investigates the forecasting ability of three different Generalised Autoregressive Conditional Heteroscedasticity (GARCH) models and the Kalman filter method. The three GARCH models applied are: bivariate GARCH, BEKK GARCH, and GARCH-GJR. Forecast errors based on 20 UK company's weekly stock return (based on time-varying beta) forecasts are employed to evaluate the out-of-sample forecasting ability of both the GARCH models and the Kalman method. Measures of forecast errors overwhelmingly support the Kalman filter approach. Among the GARCH models, GJR appears to provide somewhat more accurate forecasts than the two other GARCH models.     

Estimating stochastic volatility models using integrated nested Laplace approximations

Volatility in financial time series is mainly analysed through two classes of models; the generalized autoregressive conditional heteroscedasticity (GARCH) models and the stochastic volatility (SV) ones. GARCH models are straightforward to estimate using maximum-likelihood techniques, while SV models require more complex inferential and computational tools, such as Markov Chain Monte Carlo (MCMC). Hence, although provided with a series of theoretical advantages, SV models are in practice much less popular than GARCH ones. In this paper, we solve the problem of inference for some SV models by applying a new inferential tool, integrated nested Laplace approximations (INLAs). INLA substitutes MCMC simulations with accurate deterministic approximations, making a full Bayesian analysis of many kinds of SV models extremely fast and accurate. Our hope is that the use of INLA will help SV models to become more appealing to the financial industry, where, due to their complexity, they are rarely used in practice.     

American option pricing with discrete and continuous time models: An empirical comparison

This paper considers discrete time GARCH and continuous time SV models and uses these for American option pricing. We first of all show that with a particular choice of framework the parameters of the SV models can be estimated using simple maximum likelihood techniques. We then perform a Monte Carlo study to examine their differences in terms of option pricing, and we study the convergence of the discrete time option prices to their implied continuous time values. Finally, a large scale empirical analysis using individual stock options and options on an index is performed comparing the estimated prices from discrete time models to the corresponding continuous time model prices. The results show that, while the overall differences in performance are small, for the in the money put options on individual stocks the continuous time SV models do generally perform better than the discrete time GARCH specifications.     

Forecasting volatility of crude oil futures using a GARCH–RNN hybrid approach

Volatility is an important element for various financial instruments owing to its ability to measure the risk and reward value of a given financial asset. Owing to its importance, forecasting volatility has become a critical task in financial forecasting. In this paper, we propose a suite of hybrid models for forecasting volatility of crude oil under different forecasting horizons. Specifically, we combine the parameters of generalized autoregressive conditional heteroscedasticity (GARCH) and Glosten–Jagannathan–Runkle (GJR)-GARCH with long short-term memory (LSTM) to create three new forecasting models named GARCH–LSTM, GJR-LSTM, and GARCH-GJRGARCH LSTM in order to forecast crude oil volatility of West Texas Intermediate on different forecasting horizons and compare their performance with the classical volatility forecasting models. Specifically, we compare the performances against existing methodologies of forecasting volatility such as GARCH and found that the proposed hybrid models improve upon the forecasting accuracy of Crude Oil: West Texas Intermediate under various forecasting horizons and perform better than GARCH and GJR-GARCH, with GG–LSTM performing the best of the three proposed models at 7-, 14-, and 21-day-ahead forecasts in terms of heteroscedasticity-adjusted mean square error and heteroscedasticity-adjusted mean absolute error, with significance testing conducted through the model confidence set showing that GG–LSTM is a strong contender for forecasting crude oil volatility under different forecasting regimes and rolling-window schemes. The contribution of the paper is that it enhances the forecasting ability of crude oil futures volatility, which is essential for trading, hedging, and purposes of arbitrage, and that the proposed model dwells upon existing literature and enhances the forecasting accuracy of crude oil volatility by fusing a neural network model with multiple econometric models.     

Option pricing based on hybrid GARCH-type models with improved ensemble empirical mode decomposition

The exploration of option pricing is of great significance to risk management and investments. One important challenge to existing research is how to describe the underlying asset price process and fluctuation features accurately. Considering the benefits of ensemble empirical mode decomposition (EEMD) in depicting the fluctuation features of financial time series, we construct an option pricing model based on the new hybrid generalized autoregressive conditional heteroskedastic (hybrid GARCH)-type functions with improved EEMD by decomposing the original return series into the high frequency, low frequency and trend terms. Using the locally risk-neutral valuation relationship (LRNVR), we obtain an equivalent martingale measure and option prices with different maturities based on Monte Carlo simulations. The empirical results indicate that this novel model can substantially capture volatility features and it performs much better than the M-GARCH and Black–Scholes models. In particular, the decomposition is consistently helpful in reducing option pricing errors, thereby proving the innovativeness and effectiveness of the hybrid GARCH option pricing model.     

Evaluating portfolio Value-at-Risk using semi-parametric GARCH models

In this paper we examine the usefulness of multivariate semi-parametric GARCH models for evaluating the Value-at-Risk (VaR) of a portfolio with arbitrary weights. We specify and estimate several alternative multivariate GARCH models for daily returns on the S&P 500 and Nasdaq indexes. Examining the within-sample VaRs of a set of given portfolios shows that the semi-parametric model performs uniformly well, while parametric models in several cases have unacceptable failure rates. Interestingly, distributional assumptions appear to have a much larger impact on the performance of the VaR estimates than the particular parametric specification chosen for the GARCH equations.     

Dynamic Asymmetric GARCH

This article develops the dynamic asymmetric GARCH (or DAGARCH)model that generalizes asymmetric GARCH models such as thatof Glosten, Jagannathan, and Runkle (GJR), introduces multiplethresholds, and makes the asymmetric effect time dependent.We provide the stationarity conditions for the DAGARCH modeland show how GJR can be obtained as a special case. Furthermore,we derive the news impact curve implied by the DAGARCH modeland demonstrate its flexibility. An application to daily stockmarket indices is presented to demonstrate the practical usefulnessof the new model.     

Multivariate GARCH Modeling of Exchange Rate Volatility Transmission in the European Monetary System

We construct a series of 3‐, 4‐ and 5‐variable multivariate GARCH models of exchange rate volatility transmission across the important European Monetary System (EMS) currencies including the French franc, the German mark, the Italian lira, and the European Currency Unit. The models are estimated without imposing the common restriction of constant correlation on both daily and weekly data from April 1979–March 1997. Our results indicate the importance of checking for specification robustness in multivariate Generalized Autoregressive Conditional Heleroskedasticity (GARCH) modeling, we find that increased temporal aggregation reduces observed volatility transmission, and that the mark plays a dominant position in terms of volatility transmission.     

Forecasting Stock Index Futures Price Volatility: Linear vs. Nonlinear Models

The study examines the relative ability of various models to forecast daily stock index futures volatility. The forecasting models that are employed range from naïve models to the relatively complex ARCH-class models. It is found that among linear models of stock index futures volatility, the autoregressive model ranks first using the RMSE and MAPE criteria. We also examine three nonlinear models. These models are GARCH-M, EGARCH, and ESTAR. We find that nonlinear GARCH models dominate linear models utilizing the RMSE and the MAPE error statistics and EGARCH appears to be the best model for forecasting stock index futures price volatility.     

A New Approach to Markov-Switching GARCH Models

The use of Markov-switching models to capture the volatilitydynamics of financial time series has grown considerably duringpast years, in part because they give rise to a plausible interpretationof nonlinearities. Nevertheless, GARCH-type models remain ubiquitousin order to allow for nonlinearities associated with time-varyingvolatility. Existing methods of combining the two approachesare unsatisfactory, as they either suffer from severe estimationdifficulties or else their dynamic properties are not well understood.In this article we present a new Markov-switching GARCH modelthat overcomes both of these problems. Dynamic properties arederived and their implications for the volatility process discussed.We argue that the disaggregation of the variance process offeredby the new model is more plausible than in the existing variants.The approach is illustrated with several exchange rate returnseries. The results suggest that a promising volatility modelis an independent switching GARCH process with a possibly skewedconditional mixture density.     

Model-free versus Model-based Volatility Prediction

The well-known ARCH/GARCH models for financial time series havebeen criticized of late for their poor performance in volatilityprediction, that is, prediction of squared returns.¹ Focusingon three representative data series, namely a foreign exchangeseries (Yen vs. Dollar), a stock index series (the S&P500index), and a stock price series (IBM), the case is made thatfinancial returns may not possess a finite fourth moment. Takingthis into account, we show how and why ARCH/GARCH models—whenproperly applied and evaluated—actually do have nontrivialpredictive validity for volatility. Furthermore, we show howa simple model-free variation on the ARCH theme can performeven better in that respect. The model-free approach is basedon a novel normalizing and variance–stabilizing transformation(NoVaS, for short) that can be seen as an alternative to parametricmodeling. Properties of this transformation are discussed, andpractical algorithms for optimizing it are given.     

Exploring Metropolitan Housing Price Volatility

This paper uses GARCH models and a panel VAR model to analyze possible time variation of the volatility of single-family home value appreciation and the interactions between the volatility and the economy, using a large quarterly data set that covers 277 MSAs in the U.S. from 1990:1 to 2002:2. We find evidence of time varying volatility in about 17% of the MSAs. Using volatility series estimated with GARCH models, we find that the volatility is Granger-caused by the home appreciation rate and GMP growth rate. On the other hand, the volatility Granger-causes the personal income growth rate but the impact is not economically significant.     

Forecasting VaR using analytic higher moments for GARCH processes

It is widely accepted that some of the most accurate Value-at-Risk (VaR) estimates are based on an appropriately specified GARCH process. But when the forecast horizon is greater than the frequency of the GARCH model, such predictions have typically required time-consuming simulations of the aggregated returns distributions. This paper shows that fast, quasi-analytic GARCH VaR calculations can be based on new formulae for the first four moments of aggregated GARCH returns. Our extensive empirical study compares the Cornish–Fisher expansion with the Johnson SU distribution for fitting distributions to analytic moments of normal and Student t, symmetric and asymmetric (GJR) GARCH processes to returns data on different financial assets, for the purpose of deriving accurate GARCH VaR forecasts over multiple horizons and significance levels.     

Time Inhomogeneous Multiple Volatility Modeling

Price variations at speculative markets exhibit positive autocorrelationand cross correlation. Due to large parameter spaces necessaryfor joint modeling of variances and covariances, multivariateparametric volatility models become easily intractable in practice.We propose an adaptive procedure that identifies periods ofsecond-order homogeneity for each moment in time. To overcomethe high dimensionality of the problem we transform the multivariateseries into a set of univariate processes. We discuss thoroughlythe implementation of the adaptive technique. Theoretical andMonte Carlo results are given. We provide two applications ofthe new method. For a bivariate exchange rate series we comparethe multivariate GARCH approach with our method and find thelatter to be more in line with the underlying assumption ofindependently distributed innovations. Analyzing a 23-dimensionalvector of asset returns we underscore the case for adaptivemodeling in high-dimensional systems.     

The Effect of Futures Market Volume on Spot Market Volatility

There has been considerable interest, both academic and regulatory, in the hypothesis that the higher is the volume in the futures market, the greater is the destabilizing effect on the stock market. We show that conventional approaches, such as adding exogenous variables to GARCH models, may lead to false inferences in tests of this question. Using a stochastic volatility model, we show that, contrary to regulatory concern and the results of other papers, contemporaneous informationless futures market trading has no significant effect on spot market volatility.     

Macroeconomic risks and REITs returns: A comparative analysis

We study the relationship between the excess returns of REITs and volatilities of macroeconomic factors in developing markets (Bulgaria and South Africa) and a ‘benchmark’ developed market (USA). As expected, our results generally indicate that conditional volatilities of macroeconomic risks, extracted through the GARCH (1,1) process, are time-varying. GARCH coefficients are largely significant for excess returns and retained principal components implying conditional time-varying volatility. We use the GMM to examine the linkage between volatilities of macroeconomic variables and REITs returns. The general result here is that macroeconomic risk cannot explain excess returns on REITs. However, we document a positive relationship between variability in REITs returns and the real economy for the US. US REITs portfolio managers and investors should be wary of fluctuations in these variables as they may accentuate volatility in REITs returns.     

Forecasting variance using stochastic volatility and GARCH

This paper estimates the conditional variance of daily Swedish OMX-index returns with stochastic volatility (SV) models and GARCH models and evaluates the in-sample performance as well as the out-of-sample forecasting ability of the models. Asymmetric as well as weekend/holiday effects are allowed for in the variance, and the assumption that errors are Gaussian is released. Evidence is found of a leverage effect and of higher variance during weekends. In both in-sample and out-of-sample comparisons SV models outperform GARCH models. However, while asymmetry, weekend/holiday effects and non-Gaussian errors are important for the in-sample fit, it is found that these factors do not contribute to enhancing the forecasting ability of the SV models.     

基于实现极差和实现波动率的中国金融市场风险测度研究

目前比较流行的金融市场风险价值研究一般采用日收益数据,并基于GARCH类模型进行估计和预测。本文利用沪深股指日内高频数据,分别通过ARFIMA模型和CARR模型对实现波动率和较新的实现极差建模,计算风险价值。通过对VaR的似然比和动态分位数等回测检验,实证分析了各种模型的VaR预测能力。结果显示,使用日内高频数据的实现波动率和实现极差模型的预测能力强于采用日数据的各种GARCH类模型。     

A Note on the Term Structure of Implied Volatilities for the Yen/U.S. Dollar Currency Option

This paper tests the relationship between short dated and long dated implied volatilities obtained from Tokyo market currency option prices by employing three different volatility models: a mean reverting model, a GARCH model, and an EGARCH model. We document evidence that long dated average expected volatility is higher than that predicted by the term structure relationship during the dramatic appreciation of yen/dollar exchange in the early 1990's. This revised version was published online in August 2006 with corrections to the Cover Date.     

Estimation of tail-related risk measures for heteroscedastic financial time series: an extreme value approach

We propose a method for estimating Value at Risk (VaR) and related risk measures describing the tail of the conditional distribution of a heteroscedastic financial return series. Our approach combines pseudo-maximum-likelihood fitting of GARCH models to estimate the current volatility and extreme value theory (EVT) for estimating the tail of the innovation distribution of the GARCH model. We use our method to estimate conditional quantiles (VaR) and conditional expected shortfalls (the expected size of a return exceeding VaR), this being an alternative measure of tail risk with better theoretical properties than the quantile. Using backtesting of historical daily return series we show that our procedure gives better 1-day estimates than methods which ignore the heavy tails of the innovations or the stochastic nature of the volatility. With the help of our fitted models we adopt a Monte Carlo approach to estimating the conditional quantiles of returns over multiple-day horizons and find that this outperforms the simple square-root-of-time scaling method.