The EM algorithm for ML Estimators under nonlinear inequalities restrictions on the parameters

One of the most powerful algorithms for obtaining maximum likelihood estimates for many incomplete-data problems is the EM algorithm. However, when the parameters satisfy a set of nonlinear restrictions, It is difficult to apply the EM algorithm directly. In this paper,we propose an asymptotic maximum likelihood estimation procedure under a set of nonlinear inequalities restrictions on the parameters, in which the EM algorithm can be used. Essentially this kind of estimation problem is a stochastic optimization problem in the M-step. We make use of methods in stochastic optimization to overcome the difficulty caused by nonlinearity in the given constraints.     

有关顺序统计量的极大似然估计

讨论了一类参数空间受样本限制的极大似然估计问题.分析了随机变量分布的非零区域与似然函数定义域的对应关系,提出如果分布的非零区域受参数限制,则无论似然方程是否可解,参数的极大似然估计必然与样本顺序统计量X_((n))或X_((1))有关,并具体分析了似然估计一定等于、一定不等于和可能等于顺序统计量X_((n))(X_((1)))的三种情形,并给出了相应的判别条件.最后分析得出在第三种判别条件之下,似然估计是否取值于x_((n))(x_((1)))视具体的样本观测值决定.     

Existence and consistency of the maximum likelihood estimator for the extreme value index

The paper is about the asymptotic properties of the maximum likelihood estimator for the extreme value index. Under the second order condition, Drees et al. [H. Drees, A. Ferreira, L. de Haan, On maximum likelihood estimation of the extreme value index, Ann. Appl. Probab. 14 (2004) 1179-1201] proved asymptotic normality for any solution of the likelihood equations (with shape parameter γ>−1/2) that is not too far off the real value. But they did not prove that there is a solution of the equations satisfying the restrictions.In this paper, the existence is proved, even for γ>−1. The proof just uses the domain of attraction condition (first order condition), not the second order condition. It is also proved that the estimator is consistent. When the second order condition is valid, following the current proof, the existence of a solution satisfying the restrictions in the above-cited reference is a direct consequence.     

Topological and non-topological exact soliton solution of the power law KdV equation

This paper obtains the exact 1-soliton solution of the perturbed Korteweg–de Vries equation with power law nonlinearity. Both topological as well as non-topological soliton solutions are obtained. The solitary wave ansatz is used to carry out this integration. The domain restrictions are identified in the process and the parameter constraints are also obtained. Finally, the numerical simulations are implemented in the paper.     

A General Vehicle Routing Problem

In this paper, we study a rich vehicle routing problem incorporating various complexities found in real-life applications. The General Vehicle Routing Problem (GVRP) is a combined load acceptance and generalised vehicle routing problem. Among the real-life requirements are time window restrictions, a heterogeneous vehicle fleet with different travel times, travel costs and capacity, multi-dimensional capacity constraints, order/vehicle compatibility constraints, orders with multiple pickup, delivery and service locations, different start and end locations for vehicles, and route restrictions for vehicles. The GVRP is highly constrained and the search space is likely to contain many solutions such that it is impossible to go from one solution to another using a single neighbourhood structure. Therefore, we propose iterative improvement approaches based on the idea of changing the neighbourhood structure during the search.     

Numerische Methoden zur Behandlung einiger Problemklassen der nichtlinearen Tschebyscheff-Approximation mit Nebenbedingungen

Summary The numerical treatment of Chebyshev approximation problems is often difficult in practice because of complicated constraints. These are common in applications, for instance in differential equations. In this paper algorithms are derived for a constructive treatment of several classes of approximation problems, including parameter restrictions by Fréchet-differentiable mappings and inequality constraints in the space of approximating functions. The convergence properties of these methods are discussed and applications to practical problems are given.     

A computational evaluation of a general branch-and-price framework for capacitated network location problems

The purpose of this paper is to illustrate a general framework for network location problems, based on column generation and branch-and-price. In particular we consider capacitated network location problems with single-source constraints. We consider several different network location models, by combining cardinality constraints, fixed costs, concentrator restrictions and regional constraints. Our general branch-and-price-based approach can be seen as a natural counterpart of the branch-and-cut-based commercial ILP solvers, with the advantage of exploiting the tightness of the lower bound provided by the set partitioning reformulation of network location problems. Branch-and-price and branch-and-cut are compared through an extensive set of experimental tests.     

Empirical likelihood hypothesis test on mean with inequality constraints

Hypothesis test on the population mean with various inequality constraints is studied in this paper.The empirical likelihood method is applied to construct test statistics.Limiting distributions of the empirical likelihood ratio test statistics are proven to be a weighted mixture of chi-square distributions.Numerical results are presented to show the validity of the proposed method.     

Improving computational capabilities for addressing volume constraints in forest harvest scheduling problems

Forest Harvest Scheduling problems incorporating area-based restrictions have been of great practical interest for several years, but only recently have advances been made that allow them to be efficiently solved. One significant development has made use of formulation strengthening using the Cluster Packing Problem. This improved formulation has allowed medium sized problems to be easily solved, but when restrictions on volume production over time are added, problem difficulty increases substantially. In this paper, we study the degrading effect of certain types of volume constraints and propose methods for reducing this effect. Developed methods include the use of constraint branching, the use of elastic constraints with dynamic penalty adjustment and a simple integer allocation heuristic. Application results are presented to illustrate the computational improvement afforded by the use of these methods.     

The restricted EM algorithm under inequality restrictions on the parameters

One of the most powerful algorithms for maximum likelihood estimation for many incomplete-data problems is the EM algorithm. The restricted EM algorithm for maximum likelihood estimation under linear restrictions on the parameters has been handled by Kim and Taylor (J. Amer. Statist. Assoc. 430 (1995) 708-716). This paper proposes an EM algorithm for maximum likelihood estimation under inequality restrictions A₀β?0, where β is the parameter vector in a linear model W=Xβ+ε and ε is an error variable distributed normally with mean zero and a known or unknown variance matrix Σ>0. Some convergence properties of the EM sequence are discussed. Furthermore, we consider the consistency of the restricted EM estimator and a related testing problem.     

Weighted quantile regression for longitudinal data using empirical likelihood

This paper proposes a new weighted quantile regression model for longitudinal data with weights chosen by empirical likelihood (EL). This approach efficiently incorporates the information from the conditional quantile restrictions to account for within-subject correlations. The resulted estimate is computationally simple and has good performance under modest or high within-subject correlation. The efficiency gain is quantified theoretically and illustrated via simulation and a real data application.     

���k-out-of-nϵͳ����Լ���²����Ĺ��ƺ��㷨

In this paper we focus on the sequential k-out-of-n model with covariates. We assume that the lifetime distribution given covariates belongs to the exponential family, and deal with log-linear model of the scale parameter of the exponential distribution. The maximum likelihood estimators (MLEs) of the model parameters with order restrictions are derived and some properties of the MLEs are discussed, and we give the algorithm of MLES and the result of simulation.     

The maximum principle for the nonlinear stochastic optimal control problem of switching systems

The aim of this paper is to present a stochastic maximum principle for an optimal control problem of switching systems. It presents necessary conditions of optimality in the form of a maximum principle for stochastic switching systems, in which the dynamic of the constituent processes takes the form of stochastic differential equations. The restrictions on transitions for the system are described through equality constraints.     

A versatile algorithm for assembly line balancing

This paper discusses a two stage graph-algorithm, which was designed to solve line balancing problems including practice relevant constraints (GALBP), such as parallel work stations and tasks, cost synergies, processing alternatives, zoning restrictions, stochastic processing times or U-shaped assembly lines. Unlike former procedures, the presented approach can be easily modified to incorporate all of the named extensions. It is not only possible to select and solve single classes of constraints, but rather any combination of them with just slight modifications.     

Effects of volume fraction condition on thermodynamic restrictions in mixture theory

Volume fraction condition is a true constraint that must be taken into consideration in deducing the thermodynamic restrictions of mixture theory applying the axiom of dissipation. For a process to be admissible, the constraints imposed by the volume fraction condition include not only the equation obtained by taking its material derivative with respect to the motion of a given phase, but also those by taking its spatial gradient. The thermodynamic restrictions are deduced under the complete constraints, the results obtained are consistent for the mixtures with or without a compressible phase, and in which the free energy of each phase depends on the densities of all phases.     

Time Optimal Path Planning for Industrial Robots: A Dynamic Programming Approach Considering Torque Derivative and Jerk Constraints

This paper presents a dynamic programming approach for calculating time optimal trajectories for industrial robots, subject to various physical constraints. In addition to path velocity, motor torque, joint velocity and acceleration constraints, the present contribution also shows how to deal with torque derivative and joint jerk limitations. First a Cartesian path for the endeffector is defined by splines using Bernstein polynomials as basis functions and is parameterized via a scalar path parameter. In order to compute the belonging quantities in configuration space, inverse kinematics is solved numerically. Using this and in combination with the dynamical model, joint torques as well as their derivatives can be constrained. For that purpose the equations of motion are calculated with the help of the Projection Equation. As a consequence of the used optimization problem formulation, the dynamical model as well as the restrictions have to be transformed to path parameter space. Due to the additional consideration of jerk and torque derivative constraints, the phase plane is expanded to a phase space. The parameterized restrictions lead to feasible regions in this space, in which the optimal solution is sought. Result of the optimization is the time behavior of the path parameter and subsequently the feed forward torques for the optimal motion on the spatial path defined by previously mentioned splines. Simulation results as well as experimental results for a three axes industrial robot are presented. (© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim)     

Scale-invariant asset pricing and consumption/portfolio choice with general attitudes toward risk and uncertainty

Motivated by notions of aversion to Knightian uncertainty, this paper develops the theory of competitive asset pricing and consumption/portfolio choice with homothetic recursive preferences that allow essentially any homothetic uncertainty averse certainty-equivalent form. The market structure is scale invariant but otherwise general, allowing any trading constraints that scale with wealth. Technicalities are minimized by assuming a finite information tree. Pricing restrictions in terms of consumption growth and market returns are derived and a simple recursive method for solving the corresponding optimal consumption/portfolio choice problem is established.     

Algebraic structure of local symmetries for two-dimensional superparticles on a curved external background

In this paper, a systematic analysis of the structure of local symmetries for(1, 0)- and(1, 1)-superparticles on a curved external background is carried out. Proceeding from the requirement of the correct number of first-and second-class constraints, the minimum set of restrictions on the background superfields is written, under which a correct inclusion of interaction is guaranteed. The most general form of local symmetries is found for models on this background. The resulting transformations involve contributions with torsion superfields, and the gauge algebra turns out to be off-shell closed and nontirivially deformed as compared to the planar case. The requirement of invariance of the action relative to the direct generalization of planar local symmetries implies the complete set of D=2 supergravity constraints on the superfields as restrictions on the background.Translated from Teoreticheskaya i Matematicheskaya Fizika, Vol. 106, No. 1, pp. 102–121, January, 1966.     

Analysis of stability and synthesis of stable control systems for a class of nonlinear nonstationary systems

On the basis of sufficient condition of the stability of linear nonstationary discrete systems stated in this paper, we carry out an analysis of stability for a special class of discrete systems with nonstationary linear part, whose parameters satisfy constraints in the form of sets, and with a scalar nonlinear function satisfying linear or nonlinear restrictions. The problem of parametric synthesis of robustly stable control systems is solved for the same class of objects. The obtained results are generalized to the class of nonstationary systems with many nonlinearities of the same type.