Maximum likelihood estimation of stable Paretian models

Stable Paretian distributions have attractive properties for empirical modeling in finance, because they include the normal distribution as a special case but can also allow for heavier tails and skewness. A major reason for the limited use of stable distributions in applied work is due to the facts that there are, in general, no closed-form expressions for its probability density function and that numerical approximations are nontrivial and computationally demanding. Therefore, Maximum Likelihood (ML) estimation of stable Paretian models is rather difficult and time consuming. Here, we study the problem of ML estimation using fast Fourier transforms to approximate the stable density functions. The performance of the ML estimation approach is investigated in a Monte Carlo study and compared to that of a widely used quantile estimator. Extensions to more general distributional models characterized by time-varying location and scale are discussed.     

Schemata, Distributions and Graphical Models in Evolutionary Optimization

In this paper the optimization of additively decomposed discrete functions is investigated. For these functions genetic algorithms have exhibited a poor performance. First the schema theory of genetic algorithms is reformulated in probability theory terms. A schema defines the structure of a marginal distribution. Then the conceptual algorithm BEDA is introduced. BEDA uses a Boltzmann distribution to generate search points. From BEDA a new algorithm, FDA, is derived. FDA uses a factorization of the distribution. The factorization captures the structure of the given function. The factorization problem is closely connected to the theory of conditional independence graphs. For the test functions considered, the performance of FDA—in number of generations till convergence—is similar to that of a genetic algorithm for the OneMax function. This result is theoretically explained.     

A numerically stable optimization method based on A homogeneous function

An improved version of an unconstrained optimization algorithm based upon a homogeneous function is presented. The method is numerically stable and uses the Bartels—Golub factorization instead of Householder's modification formula. Several numerical tests indicate that the proposed method is robust and very efficient.     

Simulation of random vectors from three-dimensional spherically symmetric stable distributions

A simple approximate algorithm of sampling random vectors from three-dimensional symmetric stable distributions based on analytical approximation of inverse distribution functions is proposed. Explicit expressions for simulating the stable distributions with indices α=0.2(0.1) 1.8 are presented. The errors of the algorithm do not exceed 1–2% in the probability density function. Proceedings of the Seminar on Stability Problems for Stochastic Models, Hajdúszoboszló, Hungary, 1997, Part III.     

Tail Behavior, Modes and other Characteristics of Stable Distributions

Stable distributions have heavy tails that are asymptotically Paretian. Accurate computations of stable densities and distribution functions are used to analyze when the Paretian tail actually appears. Implications for estimation procedures are discussed. In addition to numerically locating the mode of a general stable distribution, analytic and numeric results are given for the mode. Extensive tables of stable percentiles have been computed; aspects of these tables and the appropriateness of infinite variance stable models are discussed.     

The impact of fat tailed returns on asset allocation

This paper analyzes the asset allocation problem of an investor who can invest in equity and cash when there is time variation in expected returns on the equity. The solution methodology is multistage stochastic asset allocation problem with decision rules. The uncertainty is modeled using economic scenarios with Gaussian and stable Paretian non-Gaussian innovations. The optimal allocations under these alternative hypothesis are compared. Our computational results suggest that asset allocation may be up to 20% different depending on the utility function and the risk aversion level of the investor. Certainty equivalent return can be increased up to .13% and utility can be improved up to .72% by switching to the stable Paretian model.     

Modeling fat tails in stock returns: a multivariate stable-GARCH approach

In this paper a new multivariate volatility model is proposed. It combines the appealing properties of the stable Paretian distribution to model the heavy tails with the GARCH model to capture the volatility clustering. Returns on assets are assumed to follow a sub-Gaussian distribution, which is a particular multivariate stable distribution. In this way the characteristic function of the fitted returns has a tractable expression and the density function can be recovered by numerical methods. A multivariate GARCH structure is then adopted to model the covariance matrix of the Gaussian vectors underlying the sub-Gaussian system. The model is applied to a bivariate series of daily U.S. stock returns. Value-at-risk for long and short positions is computed and compared with the one obtained using the multivariate normal and the multivariate Student’s t distribution. Finally, exploiting the recent developments in the vast dimensional time-varying covariances modeling, possible feasible extensions of our model to higher dimensions are suggested and an illustrative example using the Dow Jones index components is presented.     

Natural gradient algorithm for stochastic distribution systems with output feedback

In this paper, we use natural gradient algorithm to control the shape of the conditional output probability density function for the stochastic distribution systems from the viewpoint of information geometry. The considered system here is of multi-input and single output with an output feedback and a stochastic noise. Based on the assumption that the probability density function of the stochastic noise is known, we obtain the conditional output probability density function whose shape is only determined by the control input vector under the condition that the output feedback is known at any sample time. The set of all the conditional output probability density functions forms a statistical manifold (M), and the control input vector and the output feedback are considered as the coordinate system. The Kullback divergence acts as the distance between the conditional output probability density function and the target probability density function. Thus, an iterative formula for the control input vector is proposed in the sense of information geometry. Meanwhile, we consider the convergence of the presented algorithm. At last, an illustrative example is utilized to demonstrate the effectiveness of the algorithm.     

A parallel algorithm to solve the stable marriage problem

In this paper a parallel algorithm to solve the stable marriage problem is given. The worst case performance of this algorithm is stated. A theoretical analysis shows that the probability of the occurrence of this worst case is extremely small. For instance, if there are sixteen men and sixteen women involved, then the probability that the worst case occurs is only 10^–45. Possible future research is also discussed in this paper.     

Solving large-scale minimax problems with the primal—dual steepest descent algorithm

This paper shows that the primal-dual steepest descent algorithm developed by Zhu and Rockafellar for large-scale extended linear—quadratic programming can be used in solving constrained minimax problems related to a generalC ² saddle function. It is proved that the algorithm converges linearly from the very beginning of the iteration if the related saddle function is strongly convex—concave uniformly and the cross elements between the convex part and the concave part of the variables in its Hessian are bounded on the feasible region. Better bounds for the asymptotic rates of convergence are also obtained. The minimax problems where the saddle function has linear cross terms between the convex part and the concave part of the variables are discussed specifically as a generalization of the extended linear—quadratic programming. Some fundamental features of these problems are laid out and analyzed.This work was supported by Eliezer Naddor Postdoctoral Fellowship in Mathematical Sciences at the Department of Mathematical Sciences, the Johns Hopkins University during the year 1991–92.     

Analysis and computation of probability density functions for a 1-D impulsively controlled diffusion process

This paper proposes appropriate boundary conditions to be equipped with Kolmogorov's Forward Equation that governs a stationary probability density function for a 1-D impulsively controlled diffusion process and derives an exact probability density function. The boundary conditions are verified numerically with a Monte Carlo approach. A finite-volume method for solving the equation is also presented and its accuracy is investigated through numerical experiments.     

Estimation of a probability density function and its derivatives

We investigate statistical estimates of a probability density distribution function and its derivatives. As the starting point of the investigation we take a priori assumptions about the degree of smoothness of the probability density to be estimated. By using these assumptions we can construct estimates of the probability density function itself and its derivatives which are distinguished by the high rate of decrease of the error in the estimate as the sample size increases.Translated from Matematicheskie Zametki, Vol. 12, No. 5, 621–626, November, 1972.     

An algorithm for calculating the frequency dependence of the complex dielectric permittivity and the complex compliance

Conclusions 1. The algorithm developed for a numerical transform of the kernel in the Boltzmann — Volterra integral equation from the time domain to the frequency domain is suitable for calculating the complex dielectric permittivity from the polarization current or for calculating the complex compliance from the creep kernel.2. The algorithm is applicable to any distribution of relaxation times and has been based on changing the readings of the given function with time according to a geometrical progression. Its use requires that a certain number of coefficients be known and that readings of the given function be taken with the corresponding time coordinates.3. Calculations using expressions where the distribution of readings of the given function is shifted with respect to the time corresponding to the selected frequency makes it possible to extend the frequency range within which the sought function (complex dielectric permittivity or complex compliance) is to be determined.Institute of Polymer Mechanics, Academy of Sciences of the Latvian SSR, Riga. Translated from Mekhanika Polimerov, No. 3, pp. 524–530, May–June, 1977.     

给定密度随机数的生成的通用方法与球面和球体上均匀随机向量的生成法

本文首先介绍基于垂直概率密度表示的,给定密度函数的随机数生成的通用方法;然后介绍球面及球体上均匀随机向量的生成算法。     

Descents, Quasi-Symmetric Functions, Robinson-Schensted for Posets, and the Chromatic Symmetric Function

We investigate an apparent hodgepodge of topics: a Robinson-Schensted algorithm for (3 + 1)-free posets, Chung and Graham's G-descent expansion of the chromatic polynomial, a quasi-symmetric expansion of the path-cycle symmetric function, and an expansion of Stanley's chromatic symmetric function X _G in terms of a new symmetric function basis. We show how the theory of P-partitions (in particular, Stanley's quasi-symmetric function expansion of the chromatic symmetric function X _G ) unifies them all, subsuming two old results and implying two new ones. Perhaps our most interesting result relates to the still-open problem of finding a Robinson-Schensted algorithm for (3 + 1)-free posets. (Magid has announced a solution but it appears to be incorrect.) We show that such an algorithm ought to respect descents, and that the best partial algorithm so far—due to Sundquist, Wagner, and West—respects descents if it avoids a certain induced subposet.     

Numerically Stable Generation of Correlation Matrices and Their Factors

Correlation matrices—symmetric positive semidefinite matrices with unit diagonal—are important in statistics and in numerical linear algebra. For simulation and testing it is desirable to be able to generate random correlation matrices with specified eigenvalues (which must be nonnegative and sum to the dimension of the matrix). A popular algorithm of Bendel and Mickey takes a matrix having the specified eigenvalues and uses a finite sequence of Givens rotations to introduce 1s on the diagonal. We give improved formulae for computing the rotations and prove that the resulting algorithm is numerically stable. We show by example that the formulae originally proposed, which are used in certain existing Fortran implementations, can lead to serious instability. We also show how to modify the algorithm to generate a rectangular matrix with columns of unit 2-norm. Such a matrix represents a correlation matrix in factored form, which can be preferable to representing the matrix itself, for example when the correlation matrix is nearly singular to working precision.     

随机变量函数概率密度的定点算法

利用曲线积分和曲面积分作为工具,导出计算随机变量函数的密度函数的一种定点算法,并借助实例说明相应计算公式的应用.     

Optimal pricing for a heterogeneous portfolio for a given risk factor and convex distance measure

Consider a portfolio containing heterogeneous risks. The premiums of the policyholders might not cover the amount of the payments which an insurance company pays the policyholders. When setting the premium, this risk has to be taken into consideration. On the other hand the premium that the insured pays has to be fair. This fairness is measured by a function of the difference between the risk and the premium paid—we call this function a distance function. For a given small probability of insolvency, we find the premium for each class, such that the distance function is minimized. Next we formulate and solve the dual problem, which is minimizing the insolvency probability under the constraint that the distance function does not exceed a given level. This paper generalizes a previous paper [Zaks, Y., Frostig, E., Levikson, B., 2006. Optimal pricing of a heterogeneous portfolio for a given risk level. Astin Bull. 36 (1), 161–185] where only a square distance function was considered.     

A primal—dual affine-scaling potential-reduction algorithm for linear programming

We propose a potential-reduction algorithm which always uses the primal—dual affine-scaling direction as a search direction. We choose a step size at each iteration of the algorithm such that the potential function does not increase, so that we can take a longer step size than the minimizing point of the potential function. We show that the algorithm is polynomial-time bounded. We also propose a low-complexity algorithm, in which the centering direction is used whenever an iterate is far from the path of centers.This paper is dedicated to Phil Wolfe on the occasion of his 65th birthday.