Calibration and Filtering for Multi Factor Commodity Models with Seasonality: Incorporating Panel Data from Futures Contracts

We construct a general multi-factor model for estimation and calibration of commodity spot prices and futures valuation. This extends the multi-factor long-short model in Schwartz and Smith (Manag Sci 893–911, 2000) and Yan (Review of Derivatives Research 5(3):251–271, 2002) in two important aspects: firstly we allow for both the long and short term dynamic factors to be mean reverting incorporating stochastic volatility factors and secondly we develop an additive structural seasonality model. In developing this non-linear continuous time stochastic model we maintain desirable model properties such as being arbitrage free and exponentially affine, thereby allowing us to derive closed form futures prices. In addition the models provide an improved capability to capture dynamics of the futures curve calibration in different commodities market conditions such as backwardation and contango. A Milstein scheme is used to provide an accurate discretized representation of the s.d.e. model. This results in a challenging non-linear non-Gaussian state-space model. To carry out inference, we develop an adaptive particle Markov chain Monte Carlo method. This methodology allows us to jointly calibrate and filter the latent processes for the long-short and volatility dynamics. This methodology is general and can be applied to the estimation and calibration of many of the other multi-factor stochastic commodity models proposed in the literature. We demonstrate the performance of our model and algorithm on both synthetic data and real data for futures contracts on crude oil.     

Option pricing under joint dynamics of interest rates, dividends, and stock prices

This paper proposes a unified framework for option pricing, which integrates the stochastic dynamics of interest rates, dividends, and stock prices under the transversality condition. Using the Vasicek model for the spot rate dynamics, I compare the framework with two existing option pricing models. The main implication is that the stochastic spot rate affects options not only directly but also via an endogenously determined dividend yield and return volatility; consequently, call prices can be decreasing with respect to interest rates.     

A forward–backward SDE approach to affine models

We consider factor models for interest rates and asset prices where the risk- neutral dynamics of the factors process is modelled by an affine diffusion. We characterize the factors process and bond price in terms of forward–backward stochastic differential equations (FBSDEs), prove an existence and uniqueness theorem which gives the solution explicitly, and characterize the bond price as an exponential affine function of the factors in a new way. Our approach unifies the results, based on stochastic flows, of Elliott and van der Hoek (Finance Stoch 5:511–525, 2001) with the approach, based on the Feynman-Kac formula, of Duffie and Kan (Math Finance 6(4):379–406, 1996), and addresses a mistake in the approach of Elliott and van der Hoek (Finance Stoch 5:511–525, 2001). We extend our results on the bond price to consider the futures and forward price of a risky asset or commodity.      

Valuation of futures options with initial margin requirements and daily price limit

The paper presents a valuation model of futures options trading at exchanges with initial margin requirements and daily price limit, and this result gives an academic guidance to design trading rules at exchanges. Unlike the leading work of Black, certain trading rules are considered so as to be more fit for practical futures markets. The paper prices futures options with initial margin requirements and daily price limit by duplicating them with the help of the theory of backward stochastic differential equations （BSDEs, for short）. Furthermore, an explicit expression of the price Of the call （or the put） futures option is given and also is shown to be the unique solution of the associated nonlinear partial differential equation.     

Stochastic Models for Oil Prices and the Pricing of Futures on Oil

Abstract

In this article, we investigate and compare the performance of various one-factor diffusion models in their ability to capture the behaviour of Brent crude oil prices. New proposed models, which have a three-quarters power in the diffusion term, are found to outperform all other popular models tested. Analytic solutions for futures prices under the new models are found and used to calibrate market prices. Results from the calibration show that one of the new three-quarters models with a mean-reverting property outperforms other popular models in fitting and forecasting futures prices.     

Ornstein–Uhlenbeck processes in Hilbert space with non-Gaussian stochastic volatility

We propose a non-Gaussian operator-valued extension of the Barndorff-Nielsen and Shephard stochastic volatility dynamics, defined as the square-root of an operator-valued Ornstein–Uhlenbeck process with Lévy noise and bounded drift. We derive conditions for the positive definiteness of the Ornstein–Uhlenbeck process, where in particular we must restrict to operator-valued Lévy processes with “non-decreasing paths”. It turns out that the volatility model allows for an explicit calculation of its characteristic function, showing an affine structure. We introduce another Hilbert space-valued Ornstein–Uhlenbeck process with Wiener noise perturbed by this class of stochastic volatility dynamics. Under a strong commutativity condition between the covariance operator of the Wiener process and the stochastic volatility, we can derive an analytical expression for the characteristic functional of the Ornstein–Uhlenbeck process perturbed by stochastic volatility if the noises are independent. The case of operator-valued compound Poisson processes as driving noise in the volatility is discussed as a particular example of interest. We apply our results to futures prices in commodity markets, where we discuss our proposed stochastic volatility model in light of ambit fields.     

Option pricing with mean reversion and stochastic volatility

Many underlying assets of option contracts, such as currencies, commodities, energy, temperature and even some stocks, exhibit both mean reversion and stochastic volatility. This paper investigates the valuation of options when the underlying asset follows a mean-reverting lognormal process with stochastic volatility. A closed-form solution is derived for European options by means of Fourier transform. The proposed model allows the option pricing formula to capture both the term structure of futures prices and the market implied volatility smile within a unified framework. A bivariate trinomial lattice approach is introduced to value path-dependent options with the proposed model. Numerical examples using European options, American options and barrier options demonstrate the use of the model and the quality of the numerical scheme.     

Simulating Financial Prices

A method is described for simulating daily financial prices using appropriate statistical models. Simulated open, high, low and close prices can be obtained. These simulated prices are required for a rigorous evaluation of trading systems. Appropriate software based on the models has been used by a futures trader to improve his trading decisions.     

Eurodollar futures pricing in log-normal interest rate models in discrete time

We demonstrate the appearance of explosions in three quantities in interest rate models with log-normally distributed rates in discrete time. (1) The expectation of the money market account in the Black, Derman, Toy model, (2) the prices of Eurodollar futures contracts in a model with log-normally distributed rates in the terminal measure and (3) the prices of Eurodollar futures contracts in the one-factor log-normal Libor market model (LMM). We derive exact upper and lower bounds on the prices and on the standard deviation of the Monte Carlo pricing of Eurodollar futures in the one factor log-normal Libor market model. These bounds explode at a non-zero value of volatility, and thus imply a limitation on the applicability of the LMM and on its Monte Carlo simulation to sufficiently low volatilities.     

Forward Variance Dynamics: Bergomi’s Model Revisited

Abstract

In this article, we propose an arbitrage-free modelling framework for the joint dynamics of forward variance along with the underlying index, which can be seen as a combination of the two approaches proposed by Bergomi. The difference between our modelling framework and the Bergomi (2008. Smile dynamics III. Risk, October, 90–96) models is mainly the ability to compute the prices of VIX futures and options by using semi-analytic formulas. Also, we can express the sensitivities of the prices of VIX futures and options with respect to the model parameters, which enables us to propose an efficient and easy calibration to the VIX futures and options. The calibrated model allows to Delta-hedge VIX options by trading in VIX futures, the corresponding hedge ratios can be computed analytically.     

A multi-stage financial hedging approach for the procurement of manufacturing materials

This paper addresses the problem of mitigating procurement risk that arises from volatile commodity prices by proposing a hedging strategy within a multi-stage time frame. The proposed multi-stage hedging strategy requires a commodity futures position to be correctly initialised and rebalanced with adequate volumes of short/long positions, so as to reduce the volatility in the total procurement cost that would otherwise be generated by varying commodity spot prices. The novelty in the approach is the introduction of the rebalancing of commodity futures position at defined intermediate stages. To obtain an efficient or near optimal multi-stage hedging strategy, a discrete-time stochastic control model (DSCM) is developed. Numerical experiments and Monte Carlo simulation are used to show that the proposed multi-stage hedging strategy compares favourably with the minimal-variance hedge and the one-stage hedge. A close-form optimal solution is also presented for the case when procurement volume and price are independent.     

Price limits and capital requirements of futures clearinghouses

Futures clearinghouses need capital to provide liquidity in case of default by clearing members. Price limits truncate observed futures prices and prevent observation of clearinghouses’ true default risk exposure. We show how to estimate the true default risk exposure from observed futures prices and model capital requirements using an option pricing model, which accounts for non-normality of and truncation in observed futures returns. We apply the model to the clearinghouse associated with the Winnipeg Commodity Exchange, compare required capital levels with actual capital levels and show that ignoring non-normality of futures returns causes overall capital requirements to be significantly underestimated.     

Technical efficiency estimation with multiple inputs and multiple outputs using regression analysis

Regression and linear programming provide the basis for popular techniques for estimating technical efficiency. Regression-based approaches are typically parametric and can be both deterministic or stochastic where the later allows for measurement error. In contrast, linear programming models are nonparametric and allow multiple inputs and outputs. The purported disadvantage of the regression-based models is the inability to allow multiple outputs without additional data on input prices. In this paper, deterministic cross-sectional and stochastic panel data regression models that allow multiple inputs and outputs are developed. Notably, technical efficiency can be estimated using regression models characterized by multiple input, multiple output environments without input price data. We provide multiple examples including a Monte Carlo analysis.     

Non-linear filtering and optimal investment under partial information for stochastic volatility models

This paper studies the question of filtering and maximizing terminal wealth from expected utility in partial information stochastic volatility models. The special feature is that the only information available to the investor is the one generated by the asset prices, and the unobservable processes will be modeled by stochastic differential equations. Using the change of measure techniques, the partial observation context can be transformed into a full information context such that coefficients depend only on past history of observed prices (filter processes). Adapting the stochastic non-linear filtering, we show that under some assumptions on the model coefficients, the estimation of the filters depend on a priori models for the trend and the stochastic volatility. Moreover, these filters satisfy a stochastic partial differential equations named “Kushner–Stratonovich equations”. Using the martingale duality approach in this partially observed incomplete model, we can characterize the value function and the optimal portfolio. The main result here is that, for power and logarithmic utility, the dual value function associated to the martingale approach can be expressed, via the dynamic programming approach, in terms of the solution to a semilinear partial differential equation which depends on the filters estimate and the volatility. We illustrate our results with some examples of stochastic volatility models popular in the financial literature.     

基于高频数据的中国国债期货价格发现能力研究

期货的价格发现能力是近几年国际学术界关注的热点问题,但目前理论界相关研究主要集中于商品期货和股指期货,尚缺乏专门针对中国国债期货价格发现方面的研究。随着中国5年期国债期货于2013年9月上市交易,深入研究中国市场结构下的国债期货价格发现能力有助于从微观视角掌握与其它期货品种内在运行规律的差异性。本文运用中国5年期国债期货上市交易后的5分钟高频数据,采取向量误差修正(VECM)模型和Granger因果关系检验等计量分析方法检验中国国债期货与现货价格之间的关系,并创新性地使用共同因子贡献法和信息份额法分析我国国债期货市场与现货市场对价格发现功能的贡献程度。结果表明,中国国债期货价格与现货价格之间存在长期协整关系。中国国债期货价格是现货价格的Granger成因,且两者之间存在单向的价格引导关系。同时通过实证得出中国国债期货市场在对价格发现的贡献程度上占主导地位的结论。     

A mean reverting process for pricing treasury bills and futures contracts

An analytically tractable, discrete-time single-factor model is developed for valuing treasury bills and futures contracts. It uses a multiplicative binomial foward process that creates neither negative nor implausibly large positive interest factors, and which can incorporate different possible degrees of mean reversion. The paper derives explicit formulae for bill prices, futures prices, their conditional variances and risk premia in a setting that relates the evolution of the term structure more closely to both model and data than do other similar works. In contrast to other term-structure constrained models, this paper emphasizes that in a one-factor model the martingale probabilities cannot be treated independently of the perturbation functions. The paper's empirical methods also differ from the customary approaches. Instead of comparing differences between model-predicted and observed prices, the paper applies ARCH methodology to test model-predicted ratios of conditional variances to risk premia. Our tests find influences exogenous to the model, but these factors do not seem capable of being explained with two-factor models using only interest rates.     

天然气期货和现货价格间的相关性研究

基于天然气期货价格与现货价格序列间具有强非线性特征,本文将GARCH模型和Copula函数思想进行结合,同时考虑了天然气期货和现货价格间的时变相关结构,构建了时变Copula(GARCH-Normal、GARCH-GED和GARCH-t)模型,利用美国纽约商品交易所(NYMEX)Henry Hub交易中心天然气期货价格和现货价格数据进行实证研究。实证结果表明:GARCH-GED模型能够准确地拟合天然气期货与现货价格时间序列;时变SJC-Copula函数能够更好的描述天然气期货价格与现货价格间的相关性;天然气期货与现货价格间的相关性不是对称的,上尾的相关性小于下尾相的相关性。     

A non-Gaussian Ornstein–Uhlenbeck model for pricing wind power futures

The recent introduction of wind power futures written on the German wind power production index has brought with it new interesting challenges in terms of modelling and pricing. Some particularities of this product are the strong seasonal component embedded in the underlying, the fact that the wind index is bounded from both above and below and also that the futures are settled against a synthetically generated spot index. Here, we consider the non-Gaussian Ornstein–Uhlenbeck type processes proposed by Barndorff-Nielsen and Shephard in the context of modelling the wind power production index. We discuss the properties of the model and estimation of the model parameters. Further, the model allows for an analytical formula for pricing wind power futures. We provide an empirical study, where the model is calibrated to 37 years of German wind power production index that is synthetically generated assuming a constant level of installed capacity. Also, based on 1 year of observed prices for wind power futures with different delivery periods, we study the market price of risk. Generally, we find a negative risk premium whose magnitude decreases as the length of the delivery period increases. To further demonstrate the benefits of our proposed model, we address the pricing of European options written on wind power futures, which can be achieved through Fourier techniques.