TL;DR: In this paper, diferent metodos of estimación for parametros of the distribucion Weibull with respect to the forma of numeros difusos are discussed.
Abstract: Los procedimientos clasicos de estimacion para los parametros de la distribucion Weibull se encuentran basados en datos precisos. Se asume usualmente que los datos observados son numeros reales precisos. Sin embargo, algunos datos recolectados podrian ser imprecisos y ser representados en la forma de numeros difusos. Por lo tanto, es necesario generalizar los metodos de estimacion estadisticos clasicos de numeros reales a numeros difusos. En este articulo, diferentes metodos de estimacion son discutidos para los parametros de la distribucion Weibull cuando los datos disponibles estan en la forma de numeros difusos. Estos incluyen la estimacion por maxima verosimilitud, la estimacion Bayesiana y el metodo de momentos. Los procedimientos de estimacion se discuten en detalle y se comparan via simulaciones de Monte Carlo en terminos de sesgos promedios y errores cuadraticos medios.
TL;DR: In this article, distinct prior distributions are derived in a Bayesian inference of the two-parameters Gamma distribution, such as Jereys, reference, MDIP, Tibshirani and an innovative prior based on the copula approach.
Abstract: In this paper distinct prior distributions are derived in a Bayesian inference of the two-parameters Gamma distribution. Noniformative priors, such as Jereys, reference, MDIP, Tibshirani and an innovative prior based on the copula approach are investigated. We show that the maximal data information prior provides in an improper posterior density and that the dierent choices of the parameter of interest lead to dierent reference priors in this case. Based on the simulated data sets, the Bayesian estimates and credible intervals for the unknown parameters are computed and the performance of the prior distributions are evaluated. The Bayesian analysis is conducted using the Markov Chain Monte Carlo (MCMC) methods to generate samples from the posterior distributions under the above priors.
TL;DR: In this article, a multi-stage process is proposed based on an Almost Ideal Demand System (AIDS) to estimate simultaneously all the stages by the Generalized Method of Moments to obtain a joint covariance matrix of parameter estimates in order to use the Delta Method for calculating the standard deviation of the long-term elasticities estimates.
Abstract: The main objective in this paper is to obtain reliable long-term and shortterm elasticities estimates of the beef demand in Colombia using quarterly data since 1998 until 2007. However, complexity on the decision process of consumption should be taken into account, since expenditure on a particular good is sequential. In the case of beef demand in Colombia, a Multi-Stage process is proposed based on an Almost Ideal Demand System (AIDS). The econometric novelty in this paper is to estimate simultaneously all the stages by the Generalized Method of Moments to obtain a joint covariance matrix of parameter estimates in order to use the Delta Method for calculating the standard deviation of the long-term elasticities estimates. Additionally, this approach allows us to get elasticity estimates in each stage, but also, total elasticities which incorporate interaction between stages. On the other hand, the short-term dynamic is handled by a simultaneous estimation of the Error Correction version of the model; therefore, Monte Carlo simulation exercises are performed to analyse the impact on beef demand because of shocks at different levels of the decision making process of consumers. The results indicate that, although the total expenditure elasticity estimate of demand for beef is 1.78 in the long-term and the expenditure elasticity estimate within the meat group is 1.07, the total short-term expenditure elasticity is merely 0.03. The smaller short-term reaction of consumers is also evidenced on price shocks; while the total own price elasticity of beef is -0.24 in the short-term, the total and within meat group long-term elasticities are 1:95 and 1:17, respectively.
TL;DR: In this paper, the authors derived the L-moments and TLmoments of the Dagum distribution in closed form, and compared them with the conventional moments and L and conventional moments, and observed that the TL-moment estimator has less bias and root mean square errors than those of L.
Abstract: Modeling income, wage, wealth, expenditure and various other social variables have always been an issue of great concern. The Dagum distribution is considered quite handy to model such type of variables. Our focus in this study is to derive the L-moments and TL-moments of this distribution in closed form. Using L & TL-moments estimators we estimate the scale parameter which represents the inequality of the income distribution from the mean income. Comparing L-moments, TL-moments and conventional moments, we observe that the TL-moment estimator has lessbias and root mean square errors than those of L and conventional estimators considered in this study. We also find that the TL-moments have smaller root mean square errors for the coefficients of variation, skewness and kurtosis. These results hold for all sample sizes we have considered in our Monte Carlo simulation study.
TL;DR: In this paper, a metodologia de regresion of minimos cuadrados parciales, for the entorno of un numero grande de covariables en un espacio euclideo and una o varias respuestas that viven una variedad curvada llamada espacio simetrico Riemanniano, is presented.
Abstract: Recientemente ha habido un aumento en el interes de analizar diferentes tipos de datos variedad-valuados, dentro de los cuales aparecen los datos de matrices simetricas definidas positivas. En muchos estudios de analisis de imagenes medicas cerebrales, es de interes principal establecer la asociacion entre un conjunto de covariables y los datos variedad-valuados que son considerados como respuesta, con el fin de caracterizar las diferencias y formas en ciertas estructuras sub-corticales. Debido a que los datos variedad-valuados no forman un espacio vectorial, no es adecuado aplicar directamente las tecnicas estadisticas clasicas, ya que ciertas operaciones sobre espacio vectoriales no estan definidas en una variedad riemanniana general. En este articulo se realiza una aplicacion de la metodologia de regresion de minimos cuadrados parciales, para el entorno de un numero grande de covariables en un espacio euclideo y una o varias respuestas que viven una variedad curvada llamada espacio simetrico Riemanniano. Para poder llevar a cabo la aplicacion de dicha tecnica se utilizan el mapa exponencial Riemanniano y el mapa log Riemanniano sobre el conjunto de matrices simetricas positivas definida, mediante los cuales se transforman los datos a un espacio vectorial en donde se pueden aplicar tecnicas estadisticas clasicas. La metodologia es evaluada por medio de un conjunto de datos simulados en donde se analiza el comportamiento de la tecnica con respecto a la regresion por componentes principales.
TL;DR: In this paper, the authors proposed a new family of distributions, the so-called proportional hazard distribution-function, whose hazard function is proportional to hf (x), which can fit data with high asymmetry or kurtosis outside the range covered by the normal, t-student and logistic distributions, among others.
Abstract: We consider an arbitrary continuous cumulative distribution function F(x) with a probability density function f(x) = dF(x)=dx and hazard function hf (x) = f(x)=[1F(x)]: We propose a new family of distributions, the so-called proportional hazard distribution-function, whose hazard function is proportional to hf (x). The new model can fit data with high asymmetry or kurtosis outside the range covered by the normal, t-student and logistic distributions, among others. We estimate the parameters by maximum likelihood, profile likelihood and the elemental percentile method. The observed and expected information matrices are determined and likelihood tests for some hypotheses of interest are also considered in the proportional hazard normal distribution. We show an application to real data, which illustrates the adequacy of the proposed model.
TL;DR: In this article, the authors derived the distribution of positive linear combination of two chi-square variables when they are correlated through a bivariate Chi-square distribution, and the graph of the density function is presented.
Abstract: The distribution of the linear combination of two chi-square variables is known if the variables are independent. In this paper, we derive the distribution of positive linear combination of two chi-square variables when they are correlated through a bivariate chi-square distribution. Some properties of the distribution, namely, the characteristic function, cumulative distribution function, raw moments, mean centered moments, coefficients of skewness and kurtosis are derived. Results match with the independent case when the variables are uncorrelated. The graph of the density function is presented.
TL;DR: The log-skew-normal alpha-power (LPSN) model is presented, which contains the LN model and LSN model as special cases and models positive data with asymmetry and kurtosis larger than the one permitted by theLN distribution.
Abstract: We present a new set of distributions for positive data based on a skewnormal alpha-power (PSN) model including a new parameter which in turn makes the log-skew-normal alpha-power (LPSN) model more flexible than both the log-normal (LN) model and log-skew-normal (LSN) model. The LPSN model contains the LN model and LSN model as special cases. Furthermore, it models positive data with asymmetry and kurtosis larger than the one permitted by the LN distribution. Precipitation data illustrates the usefulness of the LPSN model being less influenced by outliers.
TL;DR: In this paper, it was shown that the Kullback's statistic for testing equality of several correlation matrices may be considered a modified likelihood ratio statistic when sampling from multivariate normal populations.
Abstract: In this article we show that the Kullback’s statistic for testing equality of several correlation matrices may be considered a modified likelihood ratio statistic when sampling from multivariate normal populations. We derive the asymptotic null distribution of L∗ in series involving independent chisquare variables by expanding L∗ in terms of other random variables and then inverting the expansion term by term. An example is also given to exhibit the procedure to be used when testing the equality of correlation matrices using the statistic L∗.
TL;DR: In this paper, a pregunta que ha surgido is si la segregación puede be aplicada in situaciones in which el emparejamiento no is aplicable.
Abstract: La segregacion y el emparejamiento son tecnicas para estimar las componentes de varianza en modelos mixtos. Una pregunta que ha surgido es si la segregacion puede ser aplicada en situaciones en las que el emparejamiento no es aplicable. Nuestra motivacion para esta investigacion se basa en el hecho de que se quiere una respuesta a esta pregunta y se quiere explorar esta importante clase de modelos con el fin de contribuir al desarrollo de los modelos mixtos. Esto es posible utilizando la estructura algebraica de los modelos mixtos con estructura de bloques ortogonal conmutativa. Se presentan dos ejemplos que muestran que la segregacion puede ser aplicada en situaciones donde el emparejamiento no es aplicable.
TL;DR: In this article, the univariate and bivariate compound Poisson process (CPP and BCPP) were derived for clustering of events and the skewness and kurtosis of CPP were analyzed.
Abstract: The univariate and bivariate compound Poisson process (CPP and BCPP, respectively) ensure a better description than the homogeneous Poisson process for clustering of events. In this paper, new explicit representations of the moment characteristics (general, central, factorial, binomial and ordinary moments, factorial cumulants) and some covariance structures are derived for the CPP and BCPP. Then, the skewness and kurtosis of the univariate CPP are obtained for the first time and special cases of the CPP are studied in detail. Applications to two real data sets are given to illustrate the usage of these processes.
TL;DR: In this paper, the problem of inferencias sobre the parametro de vulnerabilidad = P(X>v) and the proporcion of mezcla p was considered.
Abstract: En este articulo consideramos el problema de hacer inferencias sobre el parametro de vulnerabilidad = P(X>v) y la proporcion de mezcla p cuando X es una variable aleatoria cuya distribucion es una mezcla de dos distribuciones Gumbel y v es un valor fijo y conocido. Se propone el enfoque de verosimilitud perfil para estimar estos parametros, el cual es un metodo simple, pero poderoso, para estimar por separado un parametro de interes en presencia de parametros de estorbo desconocidos. Las inferencias sobre , p o (; p) se presentan por medio de regiones de verosimilitud perfil y se pueden obtener facilmente en una computadora. Esta metodologia se ilustra mediante un problema real donde se modela el tamano de inclusiones no metalicas en el acero.
TL;DR: In this paper, the problem of diagnosis residual desde la perspectiva of the metodos de subspacios of sub-spaces is discussed, and a comparison of two estadisticos and their distribuciones asintoticas bajo the hipotesis nula is presented.
Abstract: Este articulo trata el problema de la diagnosis residual desde la perspectiva de los metodos de subespacios. Se presentan dos estadisticos y sus distribuciones asintoticas bajo la hipotesis nula. Ambos estadisticos pueden usarse con procesos univariantes o multivariantes, son flexibles y permiten contrastar separadamente las correlaciones regulares y estacionales. El comportamiento en muestras finitas de las dos propuestas se ilustra mediante simulaciones de Monte Carlo y dos ejemplos con datos reales.
TL;DR: In this paper, modelos de regresion semiparametrica and observaciones influenciales that pueden tener efectos sobre los estimadores for this modelo are examined.
Abstract: En este articulo, se consideran modelos de regresion semiparametrica y se examinan observaciones influenciales que pueden tener efectos sobre los estimadores para este modelo. Una de las formas de medir la influencia de una observacion individual es borrando la observacion en el conjunto de
TL;DR: In this article, modelos potencia alfa simetricos asimetricos bimodales censurados con el fin de ajustar datos censurado con bimmodalidad and altos niveles de sesgo and curtosis are introduced.
Abstract: Se introducen los modelos potencia alfa simetricos asimetricos bimodales censurados con el fin de ajustar datos censurados con bimodalidad y altos niveles de sesgo y curtosis. Los momentos correspondientes son calculados, se considera la estimacion maximo verosimil para los parametros del modelo y se deriva la matriz de informacion observada. Se muestra la utilidad de los modelos propuestos a traves de dos aplicaciones con datos censurados reales relacionados con la medicion de HIV-1 RNA.
TL;DR: In this article, a metodologia for modelar conjuntamente tratamientos with niveles cuantitativos medidos in el tiempo, mediante la combinacion de tecnicas de superficies de respuesta with curvas de crecimiento.
Abstract: En este articulo se propone una metodologia para modelar conjuntamente tratamientos con niveles cuantitativos medidos en el tiempo, mediante la combinacion de tecnicas de superficies de respuesta con curvas de crecimiento. Se estiman los parametros del modelo, los cuales miden el efecto en el tiempo de los factores relacionados con el modelo de superficie de respuesta de segundo orden. Estas estimaciones se realizan a traves de una transformacion que permite expresar el modelo como un modelo clasico de MANOVA; de esta manera, se expresan y juzgan las hipotesis tradicionales. En este articulo se propone una metodologia para modelar conjuntamente tratamientos con niveles cuantitativos medidos en el tiempo, mediante la combinacion de tecnicas de superficies de respuesta con curvas de crecimiento. Se estiman los parametros del modelo, los cuales miden el efecto en el tiempo de los factores relacionados con el modelo de superficie de respuesta de segundo orden. Estas estimaciones se realizan a traves de una transformacion que permite expresar el modelo como un modelo clasico de MANOVA; de esta manera, se expresan y juzgan las hipotesis tradicionales.
TL;DR: In this article, the authors present an analisisis de correspondencias intra-tablas in two dimensiones, i.e., intra-columns and intrabloques.
Abstract: Para presentar los analisis de correspondencias intra-tablas, se usan los enfoques del analisis de correspondencias con respecto a un modelo y del analisis en componentes principales ponderado. Adicionalmente, se utiliza la relacion de los analisis de correspondencias con los modelos log-lineales para entender mejor las interacciones que cada analisis de correspondencias describe. Se desarrolla de manera detallada el analisis de correspondencias interno como un analisis de correspondencias intra-tablas en dos dimensiones y se introduce el analisis de correspondencias intrabloques. Por otra parte, se resumen las representaciones superpuestas y las ayudas para la interpretacion de las graficas asociadas a la estructura de subparticiones de la tabla. Finalmente, se ilustran los procedimientos con el analisis de una tabla de contingencia construida a partir de los resultados de las pruebas de estado realizadas a los estudiantes de educacion media en Colombia en el ano 2008.
TL;DR: It is observed that simple and elicited Beta priors are superior choices (in terms of minimum variance), depending on the sample size, number of items and the sum of responses, and the Bayesian estimation provides relatively more precise estimators than the ML Estimation.
Abstract: Item Count Technique (ICT) serves the purpose of estimating the proportion of the people with stigmatizing attributes using the indirect questioning method. An improved ICT has been recently proposed in the literature (not requiring two subsamples and hence free from finding optimum subsample sizes unlike the usual ICT) in a classical framework that performs better than the usual ICT and the Warner method of Randomized Response (RR) technique. This study extends the scope of this recently proposed ICT in a Bayesian framework using dierent priors in order to derive posterior distributions, posterior means and posterior variances. The posterior means and variances are compared in order to study which prior is more helpful in updating the item count technique. Moreover, we have compared the Proposed Bayesian estimation with Maximum Likelihood (ML) estimation. We have observed that simple and elicited Beta priors are superior choices (in terms of minimum variance), depending on the sample size, number of items and the sum of responses. Also, the Bayesian estimation provides relatively more precise estimators than the ML Estimation.
TL;DR: The use of Bayesian techniques to estimate the parameters of the simplex regression supported on the implementation of some simulations and a comparison with Beta regression is presented.
Abstract: Some variables are restricted to the open interval (0; 1) and several methods have been developed to work with them under the scheme of the regression analysis. Most of research consider maximum likelihood methods and the use of Beta or Simplex distributions. This paper presents the use of Bayesian techniques to estimate the parameters of the simplex regression supported on the implementation of some simulations and a comparison with Beta regression. We consider both models with constant variance and models with heteroscedasticity.
TL;DR: In this paper, an improved version of Singh's exponential type ratio estimator has been proposed and its properties have been studied under large sample approximation under some realistic conditions, and an empirical study has been carried out to judge the merits of the suggested estimator over others.
Abstract: This article considers the problem of estimating the population variance using auxiliary information. An improved version of Singh’s exponential type ratio estimator has been proposed and its properties have been studied under large sample approximation. It is shown that the proposed exponential type ratio estimator is more ecient than that considered by the Singh estimator, conventional ratio estimator and the usual unbiased estimator under some realistic conditions. An empirical study has been carried out to judge the merits of the suggested estimator over others.