Jose-Juan López-Espín

Two-stage least squares and indirect least squares algorithms for simultaneous equations models

This paper analyzes the solution of simultaneous equations models. Efficient algorithms for the two-stage least squares method using QR-decomposition are developed and studied. The reduction of the execution time when the structure of the matrices in each equation is exploited is analyzed theoretically and experimentally. An efficient algorithm for the indirect least squares method is developed. Some techniques are used to accelerate the solution of the problem: parallel versions for multicore systems, and extensive use of the MKL library, thus obtaining efficient, portable versions of the algorithms.

A parameterized shared-memory scheme for parameterized metaheuristics

This paper presents a parameterized shared-memory scheme for parameterized metaheuristics. The use of a parameterized metaheuristic facilitates experimentation with different metaheuristics and hybridation/combinations to adapt them to the particular problem we are working with. Due to the large number of experiments necessary for the metaheuristic selection and tuning, parallelism should be used to reduce the execution time. To obtain parallel versions of the metaheuristics and to adapt them to the characteristics of the parallel system, a unified parameterized shared-memory scheme is developed. Given a particular computational system and fixed parameters for the sequential metaheuristic, the appropriate selection of parameters in the unified parallel scheme eases the development of parallel efficient metaheuristics.

Distributed User Interfaces: Specification of Essential Properties

In the last few years, the traditional concept of user interface has been changing significantly. The development of new surprising devices supporting new amazing interaction mechanisms have changed the way in which people interact with computers. In this environment of strong technological growth, the increasing use of different displays managed by several users has improved user interaction. Combining fixed displays with wearable devices allows interaction and collaboration between users when they work together in a common task. Traditional user interfaces are evolving towards “distributed” user interfaces according to the new technological advances, allowing one or more interaction elements distributed among many different platforms in order to support interaction with one or more users. This paper offers a formal view of distributed user interfaces (DUI) as a mean to understand better their essentials properties and to establish the bases for formally proving properties as correctness and coherency. The proposal has been applied to a case study.

Benchmarking and Data Envelopment Analysis. An Approach based on Metaheuristics.

Abstract Data Envelopment Analysis (DEA) is a non-parametric technique for estimating technical efficiency of a set of units. DEA also provides information on benchmarking. In this paper, we study DEA models based on closest efficient targets, which are associated with the least distance and allow inefficient units to find the easiest way to achieve the efficient frontier. In the literature these models have been solved through unsatisfactory methods related to combinatorial NP-hard problems. In this paper, the problem is approached by metaheuristic techniques. Due to the high number of restrictions of the problem, finding solutions to be used in the metaheuristic algorithm is a difficult problem. Thus, this paper analyzes and compares some heuristic algorithms to obtain solutions of the problem. Each restriction determines the design of these heuristics. Thus, the problem is considered by adding constraints one by one. In this paper, the problem is presented and studied taking into account 9 of the 14 constraints, and the solution to this new problem is an upper bound of the optimal value of the original problem.

Obtaining simultaneous equation models from a set of variables through genetic algorithms

Abstract Traditionally, Simultaneous Equation Models (SEM) have been developed by people with a wealth of experience in the particular problem represented by the model. Developing a SEM is very difficult when there is a large number of variables. It would be useful to have an algorithm which gives a satisfactory SEM according to an information criterion. Because of the huge number of SEM possible, exhaustive search methods are not well suited, so an algorithm to obtain a SEM from a set of variables has been designed. The algorithm combines genetic and greedy methods. The behaviour of the algorithm is studied, and the results of some experiments are discussed.

Message-passing two steps least square algorithms for simultaneous equations models

The solution of Simultaneous Equations Models in high performance systems is analyzed. Message-passing algorithms for the Twostage Least Squares method are developed. Algorithms are studied theoretically and experimentally. The algorithms make extensive use of basic libraries like BLAS, LAPACK, and ScaLAPACK to obtain efficient and portable versions.

Benchmarking in Data Envelopment Analysis: An Approach Based on Genetic Algorithms and Parallel Programming

Data Envelopment Analysis (DEA) is a nonparametric technique to estimate the current level of efficiency of a set of entities. DEA also provides information on how to remove inefficiency through the determination of benchmarking information. This paper is devoted to study DEA models based on closest efficient targets, which are related to the shortest projection to the production frontier and allow inefficient firms to find the easiest way to improve their performance. Usually, these models have been solved by means of unsatisfactory methods since all of them are related in some sense to a combinatorial NP-hard problem. In this paper, the problem is approached by genetic algorithms and parallel programming. In addition, to produce reasonable solutions, a particular metaheuristic is proposed and checked through some numerical instances.

Application of Genetic Algorithms to Determine Closest Targets in Data Envelopment Analysis

This paper studies the application of a genetic algorithm (GA) for determining closest efficient targets in Data Envelopment Analysis. Traditionally, this problem has been solved in the literature through unsatisfactory methods since all of them are related in some sense to a combinatorial NP-hard problem. This paper presents and studies some algorithms to be used in the creation, crossover and mutation of chromosomes in a GA, in order to obtain an efficient metaheuristic which obtains better solutions.

Heuristics and metaheuristics for accelerating the computation of simultaneous equations models through a Steiner tree

Simultaneous equations models can be solved with a variety of algorithms. Some methods use the QR-decomposition of several matrices associated to the equations in the system. To accelerate the computation of those QR-decompositions and consequently the solution of simultaneous equations models, the QR-decomposition of a matrix associated to an equation can be obtained from that of another matrix associated to another equation which contains the variables of the first equation. A Steiner tree can be used, with nodes representing the equations in the model and with edges whose associated weight is the cost of computing the QR-decomposition of an equation from that of another equation. The Steiner tree of the graph associated to the simultaneous equations model gives the order of computation of the QR-decompositions of lowest computational cost. But the number of nodes in the graph is very large, and exact methods to obtain the Steiner tree are not applicable. In this paper, the application of heuristics and metaheuristics to approach the Steiner tree of the graph associated to a simultaneous equations model is considered. A heuristic and a genetic algorithm are presented and analyzed. The quality of the tree obtained and its usability in an algorithm to solve simultaneous equations models efficiently is experimentally studied.

A Parallel Application of Matheuristics in Data Envelopment Analysis

Data Envelopment Analysis (DEA) is a non-parametric methodology for estimating technical efficiency and benchmarking. In general, it is desirable that DEA generates the efficient closest targets as benchmarks for each assessed unit. This may be achieved through the application of the Principle of Least Action. However, the mathematical models associated with this principle are based fundamentally on combinatorial NP-hard problems, difficult to be solved. For this reason, this paper uses a parallel matheuristic algorithm, where metaheuristics and exact methods work together to find optimal solutions. Several parallel schemes are used in the algorithm, being possible for them to be configured at different stages of the algorithm. The main intention is to divide the number of problems to be evaluated in equal groups, so that they are resolved in different threads. The DEA problems to be evaluated in this paper are independent of each other, an indispensable requirement for this algorithm. In addition, taking into account that the main algorithm uses exact methods to solve the mathematical problems, different optimization software has been evaluated to compare their performance when executed in parallel. The method is competitive with exact methods, obtaining fitness close to the optimum with low computational time.