Enrique Castillo

APPROXIMATING EXTREME PROBABILITIES IN RELIABILITY ANALYSES USING POLYTOPES (2008)

Enrique Castillo, Alfonso Fernández-canteli, Roberto Mínguez

In this paper a new method for calculating the probability of failure is introduced. The method is based on transforming the initial set of variables into a n-dimensional uniform random variable in...

Linear-Least-Squares Initialization of Multilayer Perceptrons Through Backpropagation of the Desired Response (2008)

Deniz Erdogmus, Oscar Fontenla-romero, Jose C. Principe, Amparo Alonso-betanzos, Enrique Castillo

Abstract—Training multilayer neural networks is typically carried out using descent techniques such as the gradient-based backpropagation (BP) of error or the quasi-Newton approaches including the...

Shear Strength Prediction using Dimensional Analysis and Functional Networks (2008)

Amparo Alonso-betanzos, Enrique Castillo, Noelia Sánchez-maroño

Abstract. This paper presents a three steps methodology for predicting the failure shear effort in concrete beams. In the first step, dimensional analysis is applied to obtain several sets of...

A New Wrapper Method for Feature Subset Selection (2008)

Noelia Sánchez-maroño, Amparo Alonso-betanzos, Enrique Castillo

Abstract. ANOVA decomposition is used as the basis for the development of a new wrapper feature subset selection method, in which functional networks are used as the induction algorithm. The...

Sensitivity Analysis in Normal Bayesian Networks (2008)

Enrique Castillo, Uffe Kjaerulff

The paper discusses the problem of sensitivity analysis in normal Bayesian networks. The algebraic structure of the conditional means and variances, as linear and quadratic functions of the...

Electricity load forecast using functional networks. Report for EUNITE 2001 Competition, 2001. Available athttp://neuron.tuke.sk/competition (2008)

Enrique Castillo, Bertha Guijarro, Amparo Alonso

Abstract. In this paper a model using a functional network was employed to approach the problem of electricity load forecasting. Functional networks are generalised neural networks, that permit the...

Journal of Machine Learning Research 7 (2006) 1159--1182 Submitted 2/05; Revised 4/06; Published 7/06 A Very Fast Learning Method for Neural Networks (2008)

Based On Sensitivity, Enrique Castillo, Yoshua Bengio

This paper introduces a learning method for two-layer feedforward neural networks based on sensitivity analysis, which uses a linear training algorithm for each of the two layers. First, random...

Sensitivity analysis in Gaussian Bayesian networks using a symbolic-numerical technique (2003)

Enrique Castillo, Uffe Kjærulff

The paper discusses the problem of sensitivity analysis in Gaussian Bayesian networks. The algebraic structure of the conditional means and variances, as rational functions involving linear and...

Conditionally Specified Distributions: An Introduction (with comments and a rejoinder by the authors) (2001)

Arnold, Barry C., Castillo, Enrique, Sarabia, José María

A bivariate distribution can sometimes be characterized completely by properties of its conditional distributions. The present article surveys available research in this area. Questions of...

Conjuntos ortogonales y métodos polares en álgebra lineal : aplicaciones al cálculo matricial, sistemas de ecuaciones, desigualdades y programación lineal (1999)

Castillo, Enrique

Contenido: Espacios vectoriales y sistemas de ecuaciones; Conceptos básicos; Conjuntos ortogonales; Cálculo matricial mediante conjuntos ortogonales; Aplicaciones de los conjuntos ortogonales;...

Sensitivity analysis in discrete Bayesian networks (1997)

Enrique Castillo, José Manuel Gutiérrez, Ali S. Hadi

The paper presents an efficient computational method for performing sensitivity analysis in discrete Bayesian networks. The method exploits the structure of conditional probabilities of a target node...

Parametric Structure of Probabilities in Bayesian Networks (1997)

Enrique Castillo, José Manuel Gutiérrez, Ali S. Hadi

Abstract. The paper presents a method for uncertainty propagation in Bayesian networks in symbolic, as opposed to numeric, form. The algebraic structure of probabilities is characterized. The prior...

A new method for efficient symbolic propagation in discrete Bayesian networks (1996)

Enrique Castillo, José Manuel Gutiérrez, Ali S. Hadi

The paper presents a new efficient method for uncertainty propagation in discrete Bayesian networks in symbolic, as opposed to numeric, form, when considering some of the probabilities of the...

A modified simulation scheme for inference in Bayesian networks (1996)

Remco R. Bouckaert, Enrique Castillo, José Manuel Gutiérrez

Inthis paper we introduce anapproximationmethod for uncertainty propagationwhich is based ona modificationof the stratified simulation. The method uses a deterministic or perfect sample and...

Goal oriented symbolic propagation in Bayesian networks (1996)

Enrique Castillo, José Manuel Gutiérrez, Ali S. Hadi

The paper presents an efficient goal oriented algorithm for symbolic propagation in Bayesian networks. The proposed algorithm performs symbolic propagation using numerical methods. It first takes...

Condiciones necesarias para pruebas secuenciales truncadas óptimas. Hipótesis simples. (1983)

Castillo, Enrique, García, J.

The paper presents a new methodology to obtain partially sequential truncated tests which are optimum in the sense of minimizing the maximum expected sample number. This methodology is based on a...

Nuevos modelos de distribuciones de extremos basados en aproximaciones en las ramas (1983)

Castillo, Enrique, Moreno, Eladio, Puig-Pey, Jaime

En este trabajo se presenta una metodología que permite clasificar funciones de distribución absolutamente continuas unidimensionales atendiendo a sus ramas. La idea básica es que, en las ramas la...

Control de tamaños y potencias en pruebas de hipótesis (1981)

Castillo, Enrique, Luceño, Alberto, Mora Monte, Eduardo, Rodríguez, José

En este trabajo se analizan los grados de libertad de que dispone el investigador para fijar o acotar los tamaños de primera y segunda especie y las potencias garantizadas en aceptación y rechazo,...

Control de tamaños y potencias en pruebas de hipótesis (1981)

Castillo, Enrique, Luceño, Alberto, Mora Monte, Eduardo, Rodríguez, José

En este trabajo se analizan los grados de libertad de que dispone el investigador para fijar o acotar los tamaños de primera y segunda especie y las potencias garantizadas en aceptación y rechazo,...

Control de tamaños y potencias en pruebas de hipótesis (1981)

Rodríguez, José, Luceño Vázquez, Alberto, Mora, Eduardo, Castillo, Enrique

En este trabajo se analizan los grados de libertad de que dispone el investigador para fijar o acotar los tamaños de primera y segunda especie y las potencias garantizadas en aceptación y rechazo,...

Sobre la obtención de tests de potencia garantizada (1980)

Mora Monte, Eduardo, Castillo, Enrique

Este trabajo analiza y discute el planteamiento clásico de los tests de hipótesis compuesta frente a alternativa compuesta especialmente en lo que se refiere a los test uniformemente más potentes,...

Sobre la obtención de tests de potencia garantizada (1980)

Mora Monte, Eduardo, Castillo, Enrique

Este trabajo analiza y discute el planteamiento clásico de los tests de hipótesis compuesta frente a alternativa compuesta especialmente en lo que se refiere a los test uniformemente más potentes,...

Bivariate income distributions with lognormal conditionals

José Sarabia, Enrique Castillo, Marta Pascual, María Sarabia

lognormal distribution, conditionally specified models, European community household panel,

The danger model: application to a competitive market

Enrique Castillo, María Sarabia, Elena Álvarez

Competitive action, Danger model, Organizational behavior, Competitive strategy,

Trip matrix and path flow reconstruction and estimation based on plate scanning and link observations

Castillo, Enrique, Menéndez, José María, Jiménez, Pilar

This paper deals with the problem of trip matrix and path flow reconstruction and estimation based on plate scanning and link flow observations. To solve the problem, the following steps are used....

Predicting traffic flow using Bayesian networks

Castillo, Enrique, Menéndez, José María, Sánchez-Cambronero, Santos

This paper deals with the problem of predicting traffic flows and updating these predictions when information about OD pairs and/or link flows becomes available. To this end, a Bayesian network is...

Closed form expressions for choice probabilities in the Weibull case

Castillo, Enrique, Menéndez, José María, Jiménez, Pilar, Rivas, Ana

For a probabilistic discrete choice model with independent reversed Gumbel distributed random costs, closed form expressions for the choice probabilities are known under the assumption that the...

Modeling Probabilistic Networks of Discrete and Continuous Variables

Castillo, Enrique, Gutiérrez, José Manuel, Hadi, Ali S.

In this paper we show how discrete and continuous variables can be combined using parametric conditional families of distributions and how the likelihood weighting method can be used for propagating...

Multiple modes in densities with normal conditionals

Arnold, Barry C., Castillo, Enrique, Sarabia, José María, González-Vega, Laureano

Distributions with normal conditionals have biquadratic regression functions. Consequently, in contrast to classical bivariate normal distributions, their densities can be multimodal. Criteria for...

Specification of distributions by combinations of marginal and conditional distributions

Arnold, Barry C., Castillo, Enrique, Sarabia, José María

A k-dimensional density function is determined by certain combinations of marginal and conditional densities. The present paper identifies all possible such specifications. Gelman and Speed (1993)...

Multivariate normality via conditional specification

Arnold, Barry C., Castillo, Enrique, Sarabia, JoséMaría

If X is a k-dimensional random vector, we denote by X(i) the vector X with coordinate i deleted and by X(i,j) the vector X with coordinates i and j deleted. If for each i the conditional distribution...

A conditional characterization of the multivariate normal distribution

Arnold, Barry C., Castillo, Enrique, Sarabia, Jose María

If X is a k-dimensional random vector, we denote by X(i,j) the vector X with coordinates i and j deleted. If for each i, j the conditional distribution of Xi, Xj given X(i,j) = x(i,j) is classical...

Multivariate distributions with generalized Pareto conditionals

Arnold, Barry C., Castillo, Enrique, Sarabia, Jose María

Two classes of k-dimensional distributions with generalized Pareto conditionals are characterized. This subsumes and extends earlier work on distributions with Pareto conditionals.

Bivariate distributions characterized by one family of conditionals and conditional percentile or mode functions

Arnold, Barry C., Castillo, Enrique, Sarabia, José María

It is well known that full knowledge of all conditional distributions will typically serve to completely characterize a bivariate distribution. Partial knowledge will often suffice. For example,...

A Class of Conjugate Priors for Log-Normal Claims Based on Conditional Specification

José María Sarabia, Enrique Castillo, Emilio Gómez-Déniz

In this article, a new methodology for obtaining a premium based on a broad class of conjugate prior distributions, assuming lognormal claims, is presented. The new class of prior distributions arise...

Combined regression models

Enrique Castillo, Carmen Castillo, Ali Hadi, José Sarabia

Duality, Dual variables, Least absolute value, Least squares, Local influence, Mathematical programming, Minimax, Optimization problems,

An Exponential Family of Lorenz Curves

José-María Sarabia, Enrique Castillo, Daniel J. Slottje

A new method for building parametric-functional families of Lorenz curves, generated from an initial Lorenz curve (which satisfies some regularity conditions), is presented. The method is applied to...

Multivariate order statistics via multivariate concomitants

Arnold, Barry C., Castillo, Enrique, Sarabia, Jos Mara

Let denote a set of n independent identically distributed k-dimensional absolutely continuous random variables. A general class of complete orderings of such random vectors is supplied by viewing...

On multivariate order statistics. Application to ranked set sampling

Arnold, Barry C., Castillo, Enrique, María Sarabia, José

Two new concepts of order statistics for multivariate samples are introduced. In one of the versions it turns out that not every multivariate order statistic is present in every sample. These order...