Carlos N. Bouza

Carlos N. Bouza
University of Havana · Departamento de Matemática Aplicada

professor doctor

About

442
Publications
391,075
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
714
Citations
Additional affiliations
April 2018 - May 2018
Hasselt University
Position
  • Professor
Description
  • development of activities within the project BData
April 2018 - April 2018
University of Bergen
Position
  • Professor
December 2017 - December 2017
Universidad Tecnológica de Panamá
Position
  • Professor
Description
  • course & advising phd
Education
September 1966 - January 1971
University of Havana
Field of study
  • mathematics, statititics

Publications

Publications (442)
Article
Full-text available
Resumen Se realiza un estudio de cómo los turistas que acuden a Acapulco valoran las categorías autorrealización y autoconcepto. Estas juegan un papel importante en la valoración del uso del tiempo libre y la satisfacción. Este estudio fija la importancia de indicadores y su interrelación a partir de estudiar los datos suministrados por una muestra...
Article
Full-text available
ACAPULCO’S TOURISM MARKET: TOURISTS PROFILE ANALYSIS The profile of the tourists visiting Acapulco is made. This study uses a sample of 399 tourists. The obtained results allow the identification of the satisfaction profile of the tourism in terms of the offers of the market Keywords: Tourism market, satisfaction categories, data analysis
Article
No AbstractArticle http://dx.doi.org/10.1002/bimj.4710360113Letter to the editor http://dx.doi.org/10.1002/bimj.200810429.
Article
Full-text available
Abstract. This paper is concerned with the analyses of ranked set sampling procedures for dealing with the existence of missing observations. We consider sub-sampling the non-respondents strata. The first visit may serve for ranking accurately the observed non-respondents. Two problems are considered: the estimation of a mean and the difference of...
Article
Full-text available
ABSTRACT A proposal for determining post strata using simple information is developed. A stochastic programming with probabilistic constraints models it. Bounds for the approximation error and its expectation are developed. Their behavior is evaluated using Monte Carlo experiments.
Article
Full-text available
This paper presents a study of the clinical criteria on the persons consulting the physicians considering, that they were affected by dengue, in the state of Guerrero, México, during 2006. The classification made by the clinic was evaluated with respect to the golden rule represented by the laboratory results. All the consulted persons with the sym...
Article
Full-text available
This paper deals with the solution of the estimation of the mean using product type estimators when missing observations are present in a survey sampling. The Non Response striatum approach is considered and two estimators are developed. Imputation methods are also developed and two predictors are proposed. They are compared using Monte Carlo exper...
Article
Full-text available
ABSTRACT: In this paper we consider the effect of using an approximation to the upper bound of the Minimax optimal solution in Stochastic Linear Programming. It can be computed accepting the normality or by using a re-sampling method. Simulated Annealing is introduced for looking for the second stage solution. Least Absolute Deviation is proposed a...
Article
Full-text available
ABSTRACT This paper is devoted to the study of the estimation of the population mean using of product estimator under ranked set . Strategies for different alternative estimators are developed and compared with theirs simple random sampling with replacement counterparts. The sample errors of them are compared analytically. KEY WORDS: product estima...
Article
Full-text available
A proposal for determining post strata using simple information is developed. A stochastic programming with probabilistic constraints models it. Bounds for the approximation error and its expectation are developed. Their behavior is evaluated using Monte Carlo experiments.
Article
Full-text available
Actualmente la comercialización del agua embotellada ha cobrado importancia en la mayoría de países de América latina, realizando fuertes inversiones en su desarrollo tanto el capital transnacional como los capitales nativos. El presente trabajo presenta un estudio de mercado solicitado por los ejidatarios del poblado Azinyahualco, Municipio de Chi...
Article
Full-text available
The estimation of a diversity index is studied using the Bayesian framework provided by the superpopulation model theory. The main theoretical properties are studied when Fager and Simpson indexes are used. Different distributions are considered for illustrating the behaviour of the expected errors under the analysed models.
Article
Full-text available
Abstract: In this document the existence of a causal relationship between the promotions of sales and the diffusion of the Galician newspapers El Correo Gallego, El Progreso, El Faro de Vigo and La Región durign the period 1991-2000 is analysed . The methodology used consisted in the elaboration of regression models in which the explained variable...
Article
Full-text available
RESUMEN El problema de estimar o predecir el número de clases el que se divide una población U está presente en muchas aplicaciones y plantea varias interrogantes a la estadística. Muchas de ellas no están resueltas Good (1953) y Esty (1986) desarrollaron Teoremas Centrales del Límite. Sus hipótesis son algo fuertes. ABSTRACT: The problem of estima...
Article
Full-text available
This is the third paper of the sequel on Games Theory. In it co-operative games are presented. They assume the rationality of the players. Each player tries to win in the competition with his opponents. Neumann--Morgerstern, basically studied this model in their book. We also discuss some particular games.
Article
Full-text available
��ABSTRACT: In this paper the de-codification of labour market signals is studied. The structure of the corresponding signalling game is described. The decisions of the employer are considered as based on a classification of the aspirant. The evaluation of the posterior probability of classifying him correctly provides a decision rule. A naïve esti...
Article
Full-text available
Different methods are used for fitting the receiver operating characteristic curve of the similarity between the descriptors of the documents and the used in the search. They are compared using the search of scientific documents in a research institute. The accuracy of the procedures and the probabilities of error of each of them is computed.
Article
Full-text available
dos clases de modelos: a)modelos iid (independientes e idénticamente distribuidos), no necesariamente gaussianos y b)modelos condicionales gaussianos. Comparamos métodos paramétricos como la aproximación "varianza-covarianza", los modelos de la media móvil simple (SIM), la media móvil con ponderación exponencial (EWMA) y los modelos GARCH, la volat...
Article
Full-text available
This paper analyses the effect of selling promotions in the diffusion of daily journals in Galicia, specially with the newspapers: El Correo Gallego, El Progreso, El Faro de Vigo y La Región. With these purposes for each newspaper a classic linear regression model is derived and it is fitted using Ordinary Least Squares. Once the econometric estima...
Article
Full-text available
RESUMEN En este documento se analiza la existencia de una relación causal entre las promociones de ventas y la difusión de los diarios gallegos: El Correo Gallego, El Progreso, El Faro de Vigo y La Región en el periodo 1991-2000. La metodología utilizada consiste en la elaboración de modelos de regresión en el que la variable explicada es la difusi...
Conference Paper
Full-text available
Abstract: This article studies the diffusion of the one most important newspapers in Spain,it´s the most ancient, it´s called: El Faro de Vigo.First of all we do a descriptive analysis before modelling the diffusion of ElFaro de Vigo using Box-Jenkins methodology. We identify the ARIMA model whichhas the best explanation of the time serie studied....
Article
Full-text available
ABSTRACT The estimation of the index of Simpson is studied for a sampling plot. The properties of the estimator are developed for a fixed number of observed individuals. Estimators are built for m samples considering two possible structures. The design used is Simple Random Sampling with replacement. The developed estimators is unbiased. Their vari...
Article
Full-text available
ABSTRACT In many applications the importance of evaluating the consensus is growing. This is particularly important in many biomedical studies where the role of the experts is crucial. That is the case when the efficacy of treatments, the classification of patients, the quality of the biosphere in certain places or tests is studied. This paper is d...
Article
Full-text available
A biographical sketch of Prof. Dantzig, recently dead, is given. He is considered as one of the founding fathers of Mathematical Programming.
Article
Full-text available
ABSTRACT Traditionally we have used the simple random sampling to select samples. Most of the models of the statistic are supported by the use of samples selected by means of this design. During the last decade it has been begun to be used an alternative design that shows an improvement, in some cases, with respect to the sampling error and the acc...
Article
Full-text available
�� ABSTRACT: The estimation of the difference of means of two variables is studied when missing observations are present. The model follows the approach proposed in the paper of Bouza and Prabhu-Ajgaonkar (1993, Biometrical J., v.35, p.245-252) when simple random sampling is the design used for selecting samples and sub-samples. In this paper, a ra...
Article
Full-text available
Afrontamos aquí el importante reto de la estimación y control del riesgo de mercado, originado por cambios en los precios de acciones, mercancías, tantos de cambio y tantos de interés. Analizamos los modelos internos del Comité de Basilea para la estimación del riesgo de mercado, y se trata de hacer frente al incremento de la volatilidad de los act...
Article
Full-text available
RESUMEN Se estudia la estimación del índice de Simpson para un sitio de muestreo. Las propiedades del estimador son desarrolladas condicionadas al número de individuos observados. Se construyen estimadores para m muestras considerando dos posibles estructuras. El diseño utilizado es el muestreo simple aleatorio con reemplazo. Los estimadores desarr...
Article
Full-text available
Un estudio de la ineficiencia del ajuste mínimo cuadrático respecto al de de mínima suma de desviaciones absolutas es efectuado. Se analizan los comportamientos de estos criterios al trabajar con muestras censuradas o con el problema semi-lineal, con un proceso ARMA o con el problema multivariado. Cada análisis comparativo se completa utilizando un...
Article
Full-text available
The determination of the optimum lot size is a stochastic problem because of the randomness of the demands. The usual approaches consider that the involved distributions are known. We consider the case in which they are unknown. The optimization problem is probabilistic constraint program. The demands are modeled by an autoregressive process and th...
Article
Full-text available
��ABSTRACT: The consistency of three classes of estimators is analyzed. Two of them are based on delta sequences. The other uses record-breaking data. The conditions for ensuring that the corresponding estimator converges to the true density are discussed. A Monte Carlo experiment is developed for evaluating the behavior of some estimators ��KEYWOR...
Article
Full-text available
ABSTRACT A new proof of the Strong Consistency of a non parametric estimator of the quantiles is provided. The estimate uses record breaking data. The proposition uses a smaller number of hypothesis than other similar. The application to some well known problems as the determination of Optimal-burn-in-time, pollution and the newsboy vendor problem...
Article
Full-text available
Traditional portfolio optimization uses the standard deviation of the returns as a measure of risk. In recent years the Target-Shortfall-Probability (TSP) was discussed as an alternative measure. From the utility theoretical point of view the TSDP is not perfect. Furthermore it is criticized due to the insufficient description of the risk. The adva...
Article
Full-text available
ABSTRACT Commonly audit risks are studied by using a sample of items from an account. The estimate of the Total Auditing Error is the objective of the inquiry. An Eclectic Bayesian approach is used as an alternative to popular methods as Dollar Unit Sampling. Jacknife is used for computing the standard deviation of one of the predictors. The proced...
Article
In this paper the estimation of biodiversity is developed by deriving an estimator of an index. It is considered as a parameter and should be estimated when a simple random sample of plots is selected. Its variances is obtained. Pest dynamics is studied using a Data Mining tool. We used data generated by a large study on the behavior of a sugar can...
Article
Full-text available
The behavior of estimations of the optimal inventory level is analyzed. Two models are studied. The demands follow unknown probability distribution function. The included density functions are estimated and a plug-in rule is suggested for computing estimates of the optimal levels. Two search algorithms are proposed and compared using Monte Carlo ex...
Chapter
Full-text available
Se modela como los expertos pueden introducir infomacion e los modleso decionalaes
Article
Full-text available
The estimation of the population mean when ranked set sampling [rss] is used for selecting the sample and non responses [nr] are present, is studied. The nr stratum is sub sampled using simple random sampling with replacement. Two strategies are analyzed. One of them is based on the selection of a sub sample from the nr in each cycle. The other use...
Article
The existence of missing observations when the difference of means is estimated determines the need of sub sampling among the non respondents. Ranked set sampling is used for sub sampling. The information provided on one of the variables by the non respondents at the first attempt permits to rank them. The behavior of a ranked set sampling model wi...
Article
Full-text available
El estudio de un juego puede ser modelado asumiendo que solo algunas partidas son observadas. Entonces el árbol del juego debe ser estimado utilizando información muestral. Similarmente ocurre al obtener información sobre el comportamiento de las decisiones tomadas por individuos muestreados sobre un árbol de decisión teórico. Al considerar una med...
Article
Full-text available
Hemi-field visual-spatial attention was studied on a timed letter cancellation task in proficient readers from 7th to 9th grades. Participants maintained greater accuracy in identifying targets in the left quadrants (top/bottom) when compared with the right quadrants (top/bottom), revealing a leftward bias in their responses. Effects of gender and...
Article
Full-text available
Summary The ability in diagnosis of a separator variable is studied using a superpopulation model for stratified sampling. The estimated errors of the predictors and the estimators are similar. The model errors are computed easily. They averaged smaller computing time in the analyzed examples.
Article
Full-text available
En este trabajo analizamos el desarrollo y los conceptos de las normas L1 y L2 y se comparan con algunos ejemplos. Por una parte la norma L1 es óptima bajo los supuestos de que los errores tienen la distribución de Laplace. Esta fue propuesta mucho antes que el MC, pero las facilidades de cómputo de éstos últimos le dieron primacía. Actualmente va...
Article
The existence of missing observations when the difference of means is estimated determines the need of sub sampling among the non respondents. Ranked set sampling is used for sub sampling. The information provided on one of the variables by the non respondents at the first attempt permits to rank them. The behavior of a ranked set sampling model wi...
Article
Full-text available
The problem of Ranked Set Sampling [RSS] is analyzed and a ratio estimator is proposed. A simple linear regression superpopulation model is proposed as a counterpart to the design approach. A predictor is developed by using shrinkage techniques.
Article
Full-text available
ABSTRACT Ranked set sampling permits to use anciliary information to rank the units. The drawing of them using the ranks is at least equivalent to random sampling. In this paper this method is applied for deriving estimators of the difference. Key words: ranked set sampling, accuracy, missing observations.
Article
Full-text available
The problem of estimating the population mean under non response is studied when Ranked Set Sampling [rss] is the sampling design used for selecting the sample among the non respondents [nr]. Two rss strategies are proposed in this paper: 1. Selection of a sub sample from the nr in each cycle. 2. Selection of sub samples from the nr in each rank. T...
Article
Full-text available
A Letter Cancellation task from the Neuropsychological Diagnostic Battery of the University of Habana (DNUH-2000-R) was studied on 792 children and shown to be an appropriate diagnostic tool in classifying subjects according to sex, age, and grade level in the represented population. A method of response analysis is introduced with the purpose of d...
Article
Full-text available
Many financial models deal with concepts that are linked to regression. The unpopularity of its use in application is due to the fact that the residuals distribution is not normal. A key example is that of the study of the risk-adjusted return of the portfolio. The general equation is regarded as r - Rf = +  ( Km - Rf ) (1) Where r is the fund's...
Article
Full-text available
ABSTRACT Strategies are studied under the hyphotesis of random connections among the units. We model the case in which these conections are random. They fix networks and the sample size varies as a function of the contatcs. Hence an additional randomization stage is present. Estimators and their errors are developed. This problem is studied for sim...
Article
Full-text available
The realibility of finite systems is studied. The functioning of a system is characterized by a vector of binary variables. The a priori distribution of the random vector permits to establish a linear probability model. It is used for deriving a predictor of the realibility and its error. Under a set of mild conditions T- Student based inferences c...
Article
Full-text available
The realibility of finite systems is studied. The functioning of a system is characterized by a vector of binary variables. The a priori distribution of the random vector permits to establish a linear probability model. It is used for deriving a predictor of the realibility and its error. Under a set of mild conditions T-Student based inferences ca...
Article
Full-text available
A finite population is studied in which a random mechanism generates the coordinates of the universal parametric vector Y. A linear model M links Y i with a known variable X i . Each member i of a set of K decision makers fixes an a prori density function f i for the residuals. The true density f should be close to some member of the class to which...
Article
Full-text available
Bounds for the approximation error of optimal inventory policies are derived. The unknown distribution functions are approximated by normal distributions. The expectation of the errors are of order 0[log(N) 1 2 ]: 1 Introduction The present trend towards the introduction of new techniques in the production increases the need of setting optimal inve...
Article
Full-text available
The effect of using linear regression in a learning procedure is evaluated. A parametric programming approach is used for fitting the regression equation. The decision maker can establish an adequate model for each observed unit. His/her learning is tested by analyzing the effectivity of the adjustments.
Article
Full-text available
A number of statistical procedures for making inferences from complex sampling design are developed under superpopulation models. They permit to study the load research of electric companies. These models are usable in interactive Decision Support Systems.
Article
Full-text available
Robust estimation of the multiple regression is modeled by using a convex combination of Least Squares and Least Absolute Value criterions. A Bicriterion Parametric algorithm is developed for computing the corresponding estimates. The proposed procedure should be specially useful when outliers are expected. Its behavior is analyzed using some examp...
Article
Full-text available
A superpopulation model generates the probabilities of a Bernouilli random variable. The ranks of the involved variables are considered as survey weights. The distribution f each linear rank statistic is derived under the null hypothesis for the two sample problem and for the case k2 when a simple random sampling or stratified sampling is used. The...
Article
Full-text available
The calibration problem is analyzed using the fact t|¡¿t the current '. estimator is a ratio. The finite and the infinif.e cases are studied. . q i. Robust estimation methods perrnit to fix confidence intervals for the : . . - - ' calibrated variable under mild conditions.
Article
Full-text available
Summary The unavailability of information from some sampled units complicates the use of current sampling methods. The classical solution is to select a subsample from within the non-respondents. A random non-response mechanism provides an alternative model within the randomization approach. A superpopulation point of view suggests use of a predic...
Article
The effect of assuming an inadequate superpopulation model in stratified sampling is studied when a shrinkage technique is used. The robustness is close related with the balance of the strata samples. This theoretical problem is present in studies of ecology. The results obtained enable to estimate the abundance of plankton when salinity is the aux...
Article
Sampling strategies for estimating difference of population means are proposed. The original sample is fragmented into three subsamples. The use of a purposive device permits to measure only one of the variables in two samples. Bootstrap procedures are employed to evaluate distribution of the errors of estimators.
Article
Full-text available
We consider the modified ratio estimator proposed by Swain (2013). It is an extension of the classic ratio estimator. The objective of this paper is developing the ranked set counterpart of the new estimator, developed by Swain (2013). We derived a new ranked set sampling ratio estimator. For illustrating the behavior of the proposal a comparison o...
Article
Full-text available
The feasibili.ty of using Bootstrap Methods in the study of the approximation error in Stochastic Linear Proqramrning with Fixed Recourse is analyzed. A -*á : bound of the expected approximation error is derived by obtaining the , corresponding development in Ecigeworth Series of the statistics involved.. The convergence of the method is justified....
Article
Full-text available
Two algorithms for balancing samples are proposed. They permit to disminish the errors of a predictor. A simple linear regression superpopulation model generates the responses. The behavior of the predictor under two balancing algorithms is evaluated in a Montecarlo experiment. The data arose from the determination of the content of minerals and vi...
Article
Full-text available
The bounds of the approximation error in linear stochastic programming with completely fixed recourse are studied. Using the L 1 norm and some hypotheses on the density function estimator, a bound for the variance of the approximation error is derived. The possibility of obtaining confidence intervals for the expected error is analyzed in a jackkni...
Article
Full-text available
The eye-estimation method is widely used in practice. Several agronomic and biological measures are currently estimated by this method. If a simple linear regression is the kernel model a shrinkage technique can be used for correcting the bias associated with this method. Two predictors of the population total are proposed and the corresponding mod...