Über die Existenz überall trennscharfer Tests im nicht-dominierten Fall.
Über ein nichtparametrisches Schätzproblem.
Un teorema de convergencia con aplicación a la inferencia bayesiana.
A theorem is proved showing that, assuming some boundary conditions, the following hypotheses:1. {Xn} is a sequence of continuous random variables which approaches in probability to a numerical sequence {an},2. {Yn} is another sequence of random variables such that, for all n, the density function of Yn is proportional to the product of the density of Xn by another density not depending on n,lead to the fact that the random sequence {Yn} also approaches in probability to {an}.We also show some related...
Una alternativa bayesiana a los contrastes de la bondad del ajuste.
Se considera el problema de hacer inferencias acerca del modelo beta simétrico, desde un punto de vista bayesiano. Los resultados se aplican posteriormente al contraste de la bondad de ajuste en el caso de una hipótesis nula simple (sin parámetros marginales) y en el caso de que la hipótesis nula conste de un número finito de modelos. Caso de que el test se acepte, se dan expresiones para las probabilidades a posteriori de los diferentes modelos. En el caso de que se rechace la hipótesis nula el...
Una definición de robustez cualitativa en inferencia bayesiana.
Proponemos una definición de "estabilidad" para un modelo bayesiano, respecto a cambios en la distribución básica (la distribución a priori se mantiene fija). Esta definición se discute e interpreta mediante el concepto de continuidad.Se obtienen también condiciones suficientes para la estabilidad así definida y se proponen algunos ejemplos.Finalmente se sugiere una generalización, con la distribución a priori también variable, y se obtiene un teorema para esta nueva situación.
Una familia de distribuciones conjugadas para un proceso ARE (1).
En este artículo se estudia, desde una perspectiva bayesiana, un proceso AR(1) con errores exponenciales, ARE(1): para ello se construye una nueva familia de distribuciones conjugadas, denotada por CDG, que permite construir una especie de filtro de Kalman para la estimación recursiva de los parámetros del modelo.
Una solución bayesiana a la paradoja de Stein.
Stein (1959), en su artículo sobre la gran discrepancia entre intervalos de confianza e intervalos fiduciales, puso de relieve el comportamiento poco convincente de la distribución inicial uniforme para el vector de medias de una normal mutivariante si nuestro interés se centra en hacer inferencias sobre el cuadrado de su norma.En este artículo, utilizando el método de la maximización de la información desconocida (Bernardo, 1979), se estudia en qué sentido la distribución inicial del vector de...
Unbiased estimation for two-parameter exponential distribution under time censored sampling
The problem considered is that of unbiased estimation for a two-parameter exponential distribution under time censored sampling. We obtain a necessary form of an unbiasedly estimable parametric function and prove that there does not exist any unbiased estimator of the parameters and the mean of the distribution. For reliability estimation at a specified time point, we give a necessary and sufficient condition for the existence of an unbiased estimator and suggest an unbiased estimator based on a...
Unbiased estimation of reliability for two-parameter exponential distribution under time censored sampling
The problem considered is that of unbiased estimation of reliability for a two-parameter exponential distribution under time censored sampling. We give necessary and sufficient conditions for the existence of uniformly minimum variance unbiased estimator and also provide a characterization of a complete class of unbiased estimators in situations where unbiased estimators exist.
Unbiased estimators of multivariate discrete distributions and chi-square goodness-of-fit test.
We consider the problem of estimation of the value of a real-valued function u(θ), θ = (θ1, ..., θk)T, on the basis of a sample from non-truncated or truncated multivariate Modified Power Series Distributions. Using the general theory of estimation and the results of Patil (1965) and Patel (1978) we give the tables of MVUE's for functions of parameter θ of trinomial, multinomial, negative-multinomial and left-truncated modified power series distributions. We have applied the properties of MVUE's...
Uncertainty of coordinates and looking for dispersion of GPS receiver
The aim of the paper is to show some possible statistical solution of the estimation of the dispersion of the GPS receiver. The presented method (based on theory of linear model with additional constraints of type I) can serve for an improvement of the accuracy of estimators of coordinates acquired from the GPS receiver.
Uncertainty quantification for data assimilation in a steady incompressible Navier-Stokes problem
The reliable and effective assimilation of measurements and numerical simulations in engineering applications involving computational fluid dynamics is an emerging problem as soon as new devices provide more data. In this paper we are mainly driven by hemodynamics applications, a field where the progressive increment of measures and numerical tools makes this problem particularly up-to-date. We adopt a Bayesian approach to the inclusion of noisy data in the incompressible steady Navier-Stokes equations...
Underparametrization in a regression model with constraints II
Underparametrization of weakly nonlinear regression models
A large number of parameters in regression models can be serious obstacle for processing and interpretation of experimental data. One way how to overcome it is an elimination of some parameters. In some cases it need not deteriorate statistical properties of estimators of useful parameters and can help to interpret them. The problem is to find conditions which enable us to decide whether such favourable situation occurs.
Ungünstige Verteilungen bei einparametrigen Parameterschätzproblemen.
Uniform Minimum Variance Unbiased Estimators for Bivariate Families.
Uniformly best linear-quadratic estimator in a special structure of the regression model.
Uniformly Most Accurate Equivariant Prediction Limit.
Unimodal contaminations in testing point null hypothesis.
The problem of testing a point null hypothesis from the Bayesian perspective is considered. The uncertainties are modelled through use of ε?contamination class with the class of contaminations including: i) All unimodal distributions and ii) All unimodal and symmetric distributions. Over these classes, the infimum of the posterior probability of the point null hypothesis is computed and compared with the p?value and a better approach than the one known is obtained.
Uniqueness of the level two Bayesian network representing a probability distribution.