Monte Carlo EM加速算法

EM算法是近年来常用的求后验众数的估计的一种数据增广算法, 但由于求出其E步中积分的显示表达式有时很困难, 甚至不可能, 限制了其应用的广泛性. 而Monte Carlo EM算法很好地解决了这个问题, 将EM算法中E步的积分用Monte Carlo模拟来有效实现, 使其适用性大大增强. 但无论是EM算法, 还是Monte Carlo EM算法, 其收敛速度都是线性的, 被缺损信息的倒数所控制, 当缺损数据的比例很高时, 收敛速度就非常缓慢. 而Newton-Raphson算法在后验众数的附近具有二次收敛速率. 本文提出Monte Carlo EM加速算法, 将Monte Carlo EM算法与Newton-Raphson算法结合, 既使得EM算法中的E步用Monte Carlo模拟得以实现, 又证明了该算法在后验众数附近具有二次收敛速度. 从而使其保留了Monte Carlo EM算法的优点, 并改进了Monte Carlo EM算法的收敛速度. 本文通过数值例子, 将Monte Carlo EM加速算法的结果与EM算法、Monte Carlo EM算法的结果进行比较, 进一步说明了Monte Carlo EM加速算法的优良性.     

资产跳跃下CM策略多期收入保证价格模拟

恒定混合策略(CM策略)多期收入保证价格是保本基金发行方采取设置止损的CM\linebreak策略作为投资策略时收取保 本费的理论依据, 其中标的资产由复合泊松过程和维纳过程共同驱动, 这一定价问题内嵌奇异期权, 蒙特卡罗模拟方法擅长处理这种高维数量金融问题. 基于风险中性测度推导出多期收入保证价格的现值表达式, 用条件蒙特卡罗推导出这一现值表达式的模拟公式. 在给定参数下分别用普通蒙特卡罗和条件蒙特卡罗计算CM策略多期收入保证价格的数值解, 结果显示两种蒙特卡罗方法均能有效计算其数值解, 之后通过给定显著性水平下的置信区间长度评价两种方法的精确度, 结果显示条件蒙特卡罗比普通蒙特卡罗有很大改进. 接着运用条件蒙特卡罗模拟研究多期收入保证价格对不同参数范围的变化情况.     

Implementations of the Monte Carlo EM Algorithm

The Monte Carlo EM (MCEM) algorithm is a modification of the EM algorithm where the expectation in the E-step is computed numerically through Monte Carlo simulations. The most exible and generally applicable approach to obtaining a Monte Carlo sample in each iteration of an MCEM algorithm is through Markov chain Monte Carlo (MCMC) routines such as the Gibbs and Metropolis–Hastings samplers. Although MCMC estimation presents a tractable solution to problems where the E-step is not available in closed form, two issues arise when implementing this MCEM routine: (1) how do we minimize the computational cost in obtaining an MCMC sample? and (2) how do we choose the Monte Carlo sample size? We address the first question through an application of importance sampling whereby samples drawn during previous EM iterations are recycled rather than running an MCMC sampler each MCEM iteration. The second question is addressed through an application of regenerative simulation. We obtain approximate independent and identical samples by subsampling the generated MCMC sample during different renewal periods. Standard central limit theorems may thus be used to gauge Monte Carlo error. In particular, we apply an automated rule for increasing the Monte Carlo sample size when the Monte Carlo error overwhelms the EM estimate at any given iteration. We illustrate our MCEM algorithm through analyses of two datasets fit by generalized linear mixed models. As a part of these applications, we demonstrate the improvement in computational cost and efficiency of our routine over alternative MCEM strategies.     

基于Copula的VsR度量与事后检验

估计VaR的传统方法有三种:协方差矩阵法、历史模拟法和蒙特仁洛模拟法。通常,文献中认为刚蒙特卡洛模拟法度量VaR有很多方面的优点。但是,本文通过实证检验发现,使用传统蒙特卡洛模拟法估计的VaR偏小,事后检验效果很不理想。本文引入Copula函数来改进传统的蒙特卡洛模拟法。Copula函数能将单个边际分布和多元联合分布联系起来,能处理非正态的边际分布,并且它度量的相关性不再局限于线性相关性。实证检验表明,基于Copula的蒙特卡罗模拟法可以更加准确地度量资产组合的VaR。     

The Optimal Error of Monte Carlo Integration

The study of optimal errors of Monte Carlo methods has gained interest in recent years. Since presently no general means are available, the investigation of model problems may help one to understand the mechanisms behind them. The author provides the optimal error for the Monte Carlo integration for input data from a ball of continuous functions. As it turns out, a slight modification of the "crude Monte Carlo method" with fixed cardinality is strictly optimal even among possibly nonlinear Monte Carlo rules with varying cardinality.     

Monte Carlo methods in statistical physics: Mathematical foundations and strategies

Monte Carlo is a versatile and frequently used tool in statistical physics and beyond. Correspondingly, the number of algorithms and variants reported in the literature is vast, and an overview is not easy to achieve. In this pedagogical review, we start by presenting the probabilistic concepts which are at the basis of the Monte Carlo method. From these concepts the relevant free parameters—which still may be adjusted—are identified. Having identified these parameters, most of the tangled mass of methods and algorithms in statistical physics Monte Carlo can be regarded as realizations of merely a handful of basic strategies which are employed in order to improve convergence of a Monte Carlo computation. Once the notations introduced are available, many of the most widely used Monte Carlo methods and algorithms can be formulated in a few lines. In such a formulation, the core ideas are exposed and possible generalizations of the methods are less obscured by the details of a particular algorithm.     

Optimization of the quasi-Monte Carlo algorithm for solving systems of linear algebraic equations

An interesting conclusion about error reduction of the modified quasi-Monte Carlo method for solving systems of linear algebraic equations is suggested. The Monte Carlo method is compared with the quasi-Monte Carlo method and its modification. The optimal choice of the parameters of the Markov chain for the modified Monte Carlo method applied to solving systems of linear equations is substantiated.     

Optimal stochastic quadrature formulas for convex functions

We study optimal stochastic (or Monte Carlo) quadrature formulas for convex functions. While nonadaptive Monte Carlo methods are not better than deterministic methods, we prove that adaptive Monte Carlo methods are much better.Supported by a Heisenberg scholarship of the DFG.     

定数截尾混合指数分布在恒加应力下的MCEM加速算法

应用Monte Carlo EM加速算法给出了混合指数分布在恒加应力水平下,在定数截尾场合的参数估计问题,并通过模拟试验说明利用Monte Carlo EM加速算法来估计混合指数分布比EM算法更有效,收敛速度更快.     

Parallel Speed-Up of Monte Carlo Methods for Global Optimization

We introduce the notion of expected hitting time to a goal as a measure of the convergence rate of a Monte Carlo optimization method. The techniques developed apply to simulated annealing, genetic algorithms, and other stochastic search schemes. The expected hitting time can itself be calculated from the more fundamental complementary hitting time distribution (CHTD) which completely characterizes a Monte Carlo method. The CHTD is asymptotically a geometric series, (1/s)/(1 − λ), characterized by two parameters, s, λ, related to the search process in a simple way. The main utility of the CHTD is in comparing Monte Carlo algorithms. In particular we show that independent, identical Monte Carlo algorithms run in parallel, IIP parallelism, and exhibit superlinear speedup. We give conditions under which this occurs and note that equally likely search is linearly sped up. Further we observe that a serial Monte Carlo search can have an infinite expected hitting time, but the same algorithm when parallelized can have a finite expected hitting time. One consequence of the observed superlinear speedup is an improved uniprocessor algorithm by the technique of in-code parallelism.     

A comparison of bootstrap and Monte-Carlo testing approaches to value-at-risk diagnosis

In this note we investigate a particular resampling scheme and Monte Carlo testing to determine critical values for two test statistics typically used for diagnosing value-at-risk models. In cases of small nominal coverage subjected to testing, the dynamic quantile test and a corresponding logit based likelihood ratio test suffer from poor convergence to the asymptotic limit distribution. In terms of empirical size both resampling and Monte Carlo approaches offer most accurate test features with the Monte Carlo technique achieving power gains if a misspecified value-at-risk model is subjected to testing.     

Efficiency of Monte Carlo EM and Simulated Maximum Likelihood in Two-Stage Hierarchical Models

Likelihood estimation in hierarchical models is often complicated by the fact that the likelihood function involves an analytically intractable integral. Numerical approximation to this integral is an option but it is generally not recommended when the integral dimension is high. An alternative approach is based on the ideas of Monte Carlo integration, which approximates the intractable integral by an empirical average based on simulations. This article investigates the efficiency of two Monte Carlo estimation methods, the Monte Carlo EM (MCEM) algorithm and simulated maximum likelihood (SML). We derive the asymptotic Monte Carlo errors of both methods and show that, even under the optimal SML importance sampling distribution, the efficiency of SML decreases rapidly (relative to that of MCEM) as the missing information about the unknown parameter increases. We illustrate our results in a simple mixed model example and perform a simulation study which shows that, compared to MCEM, SML can be extremely inefficient in practical applications.     

正贡献蒙特卡罗方法

考虑不依赖时间的粒子输运问题.不妨碍一般性,还限定问题是与能量无关的.令P=(r,Ω),其中r和Ω分别表示粒子的位置和运动方向单位矢量.用S(P)表示粒子源;φ(P)表示粒子通量;D(P)表示探测器对粒子通量的响应函数.目的是要计算如下积分效应:     

On iterative and Monte Carlo solutions of the Boltzmann equation

The two most commonly used techniques for solving the Boltzmann equation, with given boundary conditions, are first iterative equations (typically the BGK equation) and Monte Carlo methods. The present work examines the accuracy of two different iterative solutions compared with that of an advanced Monte Carlo solution for a one-dimensional shock wave in a hard sphere gas. It is found that by comparison with the Monte Carlo solution the BGK model is not as satisfactory as the other first iterative solution (Holway's) and that the BGK solution may be improved by using directional temperatures rather than a mean temperature.     

A new iterative Monte Carlo approach for inverse matrix problem

A new approach of iterative Monte Carlo algorithms for the well-known inverse matrix problem is presented and studied. The algorithms are based on a special techniques of iteration parameter choice, which allows to control the convergence of the algorithm for any column (row) of the matrix using different relaxation parameters. The choice of these parameters is controlled by a posteriori criteria for every Monte Carlo iteration. The presented Monte Carlo algorithms are implemented on a SUN Sparkstation. Numerical tests are performed for matrices of moderate in order to show how work the algorithms. The algorithms under consideration are well parallelized.     

An Efficient Monte Carlo Method for Optimal Control Problems with Uncertainty

A general framework is proposed for what we call the sensitivity derivative Monte Carlo (SDMC) solution of optimal control problems with a stochastic parameter. This method employs the residual in the first-order Taylor series expansion of the cost functional in terms of the stochastic parameter rather than the cost functional itself. A rigorous estimate is derived for the variance of the residual, and it is verified by numerical experiments involving the generalized steady-state Burgers equation with a stochastic coefficient of viscosity. Specifically, the numerical results show that for a given number of samples, the present method yields an order of magnitude higher accuracy than a conventional Monte Carlo method. In other words, the proposed variance reduction method based on sensitivity derivatives is shown to accelerate convergence of the Monte Carlo method. As the sensitivity derivatives are computed only at the mean values of the relevant parameters, the related extra cost of the proposed method is a fraction of the total time of the Monte Carlo method.     

上证50ETF期权定价有效性的研究:基于B-S-M模型和蒙特卡罗模拟

上证50ETF期权作为中国资本市场上股票期权的第一个试点产品,其定价问题尤为重要。本文分别运用B-S-M期权定价模型和蒙特卡罗模拟方法对其定价进行实证研究,分析结果表明:1)IGARCH模型比传统的GARCH模型更能较好地拟合上证50ETF的波动率;2)当模拟次数为1000时,蒙特卡罗方法的效率一致地高于B-S-M模型,并且除了对偶变量技术的拟蒙特卡罗其他模型的精确度也都高于B-S-M模型;3)B-S-M模型和蒙特卡罗模拟方法都可以较为准确地、有效地模拟出上证50ETF期权价格。这些研究将为今后期权定价模型的发展和完善提供必要的参考和指引。     

Robustness and applicability of Markov chain Monte Carlo algorithms for eigenvalue problems

In this paper we analyse applicability and robustness of Markov chain Monte Carlo algorithms for eigenvalue problems. We restrict our consideration to real symmetric matrices.

Almost Optimal Monte Carlo (MAO) algorithms for solving eigenvalue problems are formulated. Results for the structure of both – systematic and probability error are presented. It is shown that the values of both errors can be controlled independently by different algorithmic parameters. The results present how the systematic error depends on the matrix spectrum. The analysis of the probability error is presented. It shows that the close (in some sense) the matrix under consideration is to the stochastic matrix the smaller is this error. Sufficient conditions for constructing robust and interpolation Monte Carlo algorithms are obtained. For stochastic matrices an interpolation Monte Carlo algorithm is constructed.

A number of numerical tests for large symmetric dense matrices are performed in order to study experimentally the dependence of the systematic error from the structure of matrix spectrum. We also study how the probability error depends on the balancing of the matrix.     

评估系统可靠度置信下限的随机模拟方法

由已知寿命分布和测试数据的五个不同类型部件,组成一串并联系统,本文应用Monte Carlo方法,Bootstrap方法,和Double Monte Carlo方法给出该系统的可靠度置信下限.