Browsing by UCM subject "Probabilidades (Matemáticas)"
Now showing 1 - 20 of 68
Results Per Page
PublicationA bipolar knowledge representation model to improve supervised fuzzy classification algorithms.(Springer-Verlag, 2018) Villarino, Guillermo; Gomez, Daniel; Rodríguez, Juan Tinguaro; Montero, JavierMost supervised classification algorithms produce a soft score (either a probability, a fuzzy degree, a possibility, a cost, etc.) assessing the strength of the association between items and classes. After that, each item is assigned to the class with the highest soft score. In this paper, we show that this last step can be improved through alternative procedures more sensible to the available soft information. To this aim, we propose a general fuzzy bipolar approach that enables learning how to take advantage of the soft information provided by many classification algorithms in order to enhance the generalization power and accuracy of the classifiers. To show the suitability of the proposed approach, we also present some computational experiences for binary classification problems, in which its application to some well-known classifiers as random forest, classification trees and neural networks produces a statistically significant improvement in the performance of the classifiers. PublicationA formal approach to the concept of structure function.(International Institute for Advanced Studies in Systems Research and Cybernetics, 1990) Montero, Javier; Hough, Robin R.; Lasker, George E., PublicationA Mathematical Pre-Disaster Model with Uncertainty and Multiple Criteria for Facility Location and Network Fortification(MDPI, 2020-04-03) Monzón, Julia; Liberatore, Federico; Vitoriano, BegoñaDisasters have catastrophic effects on the affected population, especially in developing and underdeveloped countries. Humanitarian Logistics models can help decision-makers to efficiently and effectively warehouse and distribute emergency goods to the affected population, to reduce casualties and suffering. However, poor planning and structural damage to the transportation infrastructure could hamper these efforts and, eventually, make it impossible to reach all the affected demand centers. In this paper, a pre-disaster Humanitarian Logistics model is presented that jointly optimizes the prepositioning of aid distribution centers and the strengthening of road sections to ensure that as much affected population as possible can efficiently get help. The model is stochastic in nature and considers that the demand in the centers affected by the disaster and the state of the transportation network are random. Uncertainty is represented through scenarios representing possible disasters. The methodology is applied to a real-world case study based on the 2018 storm system that hit the Nampula Province in Mozambique. PublicationA new model for the M/G/1 retrial queve(Facultad de Ciencias Económicas y Empresariales. Decanato, 1994) Falin, Guennadi I.; Rodrigo Fernández, Antonio; Vázquez Furelos, MercedesThis article deals with a new model for the M/G/1 retrial queue. We consider the process (M(t),N(t)) where M(t) is the total number of arrivals from the last departure until time t and N(t) is the number of customers in orbit at time t. We obtain the generating function together with a recurrent formula for factorial moments in the steady state. Limit behavior under heavy traffic is also studied. We use this process to get an estimator of the parameter of retrial and obtain its accuracy by solving some linear differential equations. We also give some numerical examples. PublicationA retrial model at nonstationary regime(Facultad de Ciencias Económicas y Empresariales. Instituto Complutense de Análisis Económico (ICAE), 1994) Vázquez, MercedesIn this article we analyze a model of a retrial queueing system where customers in the orbit join a queue with FCFS discipline. We adopt a nostationary regime. We derive some probabilities using fue theory of semiregenerative processes. We obtain an integral estimator for the blocking probability. We also find lower and upper bounds for the l1 distance between the probability distributions in an M/G/1/oo retrial model and the M/G/1/ queue at stationary regime. PublicationA retrial system with constant attempts(Facultad de Ciencias Económicas y Empresariales. Instituto Complutense de Análisis Económico (ICAE), 1994) Vázquez, MercedesIn this article we study a retrial queueing system in which customers in orbit join a queue with a FIFO discipline. We use the process (M(t),N(t) where M(t) represents the number of arrivals during a elapsed service time until t while N(t) stands for the random number of customers in orbit al time t. We obtain the probabitity in the steady state in the M/M/1 case. The joint generating function in the M/G/1 case is also studied. PublicationA simulation study of a nested sequence of binomial regression models(Taylor & Francis, 2007) Pardo Llorente, Julio Ángel; Pardo Llorente, Leandro; Pardo Llorente, María del CarmenThe inference problem we consider is that of model choice from a nested sequence of binomial regression models. The approach we take is to test successively, from most general to most specific, the corresponding sequence of composite hypotheses. This approach is based on the very general class of divergence measures, the phi-divergence. An approximation to the power function of the new family of test statistics proposed is obtained. An extensive simulation study is carried out by obtaining new test statistics that are a good alternative to the traditional loglikelihood test statistic. PublicationA social Capital Index(2012-01) Manuel García, Conrado MiguelIn this paper it propose a social capital measure for individuals belonging to a social network. to do this, It uses a game theoretical approach and so it suppose that these indivuduals are involved in a cooperative TU-game modelling the economic or social interest that motivate their interest. PublicationAlgunas relaciones entre la aproximación clásica y bayesiana en el contraste de hipótesis(Universidad Complutense de Madrid, Servicio de Publicaciones, 2002) Sanz San Miguel, Luis; Gómez Villegas, Miguel ÁngelPara un problema de localización en una hipótesis nula puntual se relaciona el p-valor con la probabilidad final de la hipótesis nula, cuando la distribución inicial se escoge convenientemente. Se generalizan estos resultados al caso de distribuciones e-contaminadas. PublicationAn axiomatic approach to fuzzy multicriteria analysis.(Oxford : North-Holland, 1988) Montero, Javier; Gupta, Madan M.; Yamakawa, Takeshi PublicationApplications of Gaver-Stehfest method of inverting laplace transforms to ruin theory(Facultad de Ciencias Económicas y Empresariales. Decanato, 1997) Usábel Rodrigo, Miguel ArturoThe Stehfest-Gaver method of inverting Laplace transforms is a very useful tool in approximating non-ruin probabilities. An accuracy of 6 to 10 significant digits is obtained in every case studied (Tables 1,2 and 3) except for Log-normal claim size and large initial reserves where the accuracy of the "exact" values (using Product integration ) is not guaranteed to be more than 5 digits. The efficiency in terms of computational time is also outstanding because we onIy need to evaluate 20 times the Laplace transform of the c.dJ. of the claim size as shown in (1.6). PublicationApproximations for multivariate characteristics of classical risk ruin processes(Facultad de Ciencias Económicas y Empresariales. Decanato, 1998) Usábel Rodrigo, Miguel ArturoMultivariate characteristic of risk processes are of high interest to academic actuaries. In such modele the probability of ruin ie obtained not only considering initial reserves u but the severity of ruin y and the surplus before ruin x. This ruin probability can be expressed using an integral equation that can be efficiently solved using Gaver-Stehfest method of invertig Laplace transforms. PublicationAprendizaje basado en problemas del cálculo de probabilidades a través del entorno Moodle del campus virtual(2021-05) Alonso Sanz, Rosa; Martín Campo, Francisco Javier; Landaburu Jiménez, María Elena; Pardo Llorente, María del Carmen; Tirado Domínguez, Gregorio; Rodríguez González, Juán Tinguaro; Liberatore, FedericoMediante este trabajo se pretende innovar en el aprendizaje del alumno desarrollando nuevos materiales didácticos basados en la metodología de Aprendizaje Basado en Problemas para la docencia de Cálculo de Probabilidades tomando como herramientas de Trabajo en Grupo las proporcionadas por una plataforma virtual. PublicationAsymptotic Theory for Rotated Multivariate GARCH Models(Facultad de Ciencias Económicas y Empresariales. Instituto Complutense de Análisis Económico (ICAE), 2018) Asai, Manabu; Chang, Chia-Lin; McAleer, Michael; Laurent, PauwelsIn this paper, we derive the statistical properties of a two step approach to estimating multivariate GARCH rotated BEKK (RBEKK) models. By the definition of rotated BEKK, we estimate the unconditional covariance matrix in the first step in order to rotate observed variables to have the identity matrix for its sample covariance matrix. In the second step, we estimate the remaining parameters via maximizing the quasi-likelihood function. For this two step quasi-maximum likelihood (2sQML) estimator, we show consistency and asymptotic normality under weak conditions. While second-order moments are needed for consistency of the estimated unconditional covariance matrix, the existence of finite sixthorder moments are required for convergence of the second-order derivatives of the quasilog-likelihood function. We also show the relationship of the asymptotic distributions of the 2sQML estimator for the RBEKK model and the variance targeting (VT) QML estimator for the VT-BEKK model. Monte Carlo experiments show that the bias of the 2sQML estimator is negligible, and that the appropriateness of the diagonal specification depends on the closeness to either of the Diagonal BEKK and the Diagonal RBEKK models. PublicationBottom-Up: a New Algorithm to Generate Random Linear Extensions of a Poset(Springer Netherlands, 2018-10-09) García-Segador, P.; Miranda Menéndez, PedroIn this paper we present a new method for deriving a random linear extension of a poset. This new strategy combines Probability with Combinatorics and obtains a procedure where each minimal element of a sequence of subposets is selected via a probability distribution. The method consists in obtaining a weight vector on the elements of P, so that an element is selected with a probability proportional to its weight. From some properties on the graph of adjacent linear extensions, it is shown that the probability distribution can be obtained by solving a linear system. The number of equations involved in this system relies on the number of what we have called positioned antichains, that allows a reduced number of equations. Finally, we give some examples of the applicability of the algorithm. This procedure cannot be applied to every poset, but it is exact when it can be used. Moreover, the method is quick and easy to implement. Besides, it allows a simple way to derive the number of linear extensions of a given poset. PublicationCalculating ultimate non-ruin probabilities when claim sizes follow a generalized r-convolution distribution function(Facultad de Ciencias Económicas y Empresariales. Decanato, 1998) Usábel Rodrigo, Miguel ArturoThe non-ruin probability, for initial reserves u, in the classical can be calculated using the so-called Bromwich-Mellin inversion formula, an outstanding result from Residues Theory first introduced for these purposes by Seal(1977) for exponential claim size. We will use this technique when claim sizes follow a generalized r-convolution function distribution. Some of the most frequently used heavy-tailed distributions in actuarial science belongs to this family. Thorin(1977) or Berg(1981) proved that Pareto distributions are members of this family; so Thorin(1977) did with Log-normal distributions. PublicationComposite Likelihood Methods Based on Minimum Density Power Divergence Estimator(https://www.mdpi.com/, 2017) Castilla González, Elena María; Martín, Nirian; Pardo Llorente, Leandro; Zografos, KonstantinosIn this paper, a robust version of the Wald test statistic for composite likelihood is considered by using the composite minimum density power divergence estimator instead of the composite maximum likelihood estimator. This new family of test statistics will be called Wald-type test statistics. The problem of testing a simple and a composite null hypothesis is considered, and the robustness is studied on the basis of a simulation study. The composite minimum density power divergence estimator is also introduced, and its asymptotic properties are studied.