16 results on '"Francisco-Fernández, Mario"'
Search Results
2. Automatic detection of defective crankshafts by image analysis and supervised classification
- Author
-
Remeseiro, Beatriz, Tarrío-Saavedra, Javier, Francisco-Fernández, Mario, Penedo, Manuel G., Naya, Salvador, and Cao, Ricardo
- Published
- 2019
- Full Text
- View/download PDF
3. Excel Templates: A Helpful Tool for Teaching Statistics
- Author
-
Quintela-del-Río, Alejandro and Francisco-Fernández, Mario
- Published
- 2017
4. A flexible functional-circular regression model for analyzing temperature curves
- Author
-
Meilán-Vila, Andrea, Crujeiras, Rosa M., and Francisco-Fernández, Mario
- Subjects
Methodology (stat.ME) ,FOS: Computer and information sciences ,Physics::Atmospheric and Oceanic Physics ,Statistics - Methodology - Abstract
Changes on temperature patterns, on a local scale, are perceived by individuals as the most direct indicators of global warming and climate change. As a specific example, for an Atlantic climate location, spring and fall seasons should present a mild transition between winter and summer, and summer and winter, respectively. By observing daily temperature curves along time, being each curve attached to a certain calendar day, a regression model for these variables (temperature curve as covariate and calendar day as response) would be useful for modeling their relation for a certain period. In addition, temperature changes could be assessed by prediction and observation comparisons in the long run. Such a model is presented and studied in this work, considering a nonparametric Nadaraya-Watson-type estimator for functional covariate and circular response. The asymptotic bias and variance of this estimator, as well as its asymptotic distribution are derived. Its finite sample performance is evaluated in a simulation study and the proposal is applied to investigate a real-data set concerning temperature curves., 28 pages, 11 figures
- Published
- 2022
5. Smoothing Parameter Selection Methods for Nonparametric Regression with Spatially Correlated Errors
- Author
-
Francisco-Fernandez, Mario and Opsomer, Jean D.
- Published
- 2005
6. Two tests for heteroscedasticity in nonparametric regression
- Author
-
Francisco-Fernández, Mario and Vilar-Fernández, Juan M.
- Published
- 2009
- Full Text
- View/download PDF
7. Bandwidth selection for the local polynomial estimator under dependence: A simulation study
- Author
-
Francisco-Fernández, Mario and Vilar-Fernández, Juan M.
- Published
- 2005
- Full Text
- View/download PDF
8. Kernel distribution estimation for grouped data
- Author
-
Barreiro-Ures, Daniel, Cao, Ricardo, Francisco-Fernández, Mario, and Reyes, Miguel
- Subjects
Estadística matemàtica ,Bootstrap bandwidth ,Anàlisi de supervivència (Biometria) ,Matemàtiques i estadística::Estadística matemàtica [Àrees temàtiques de la UPC] ,interval data ,plug-in bandwidth ,cumulative distribution function estimator ,62 Statistics::62N Survival analysis and censored data [Classificació AMS] ,62 Statistics::62G Nonparametric inference [Classificació AMS] - Abstract
Interval-grouped data appear when the observations are not obtained in continuous time, but monitored in periodical time instants. In this framework, a nonparametric kernel distribution estimator is proposed and studied. The asymptotic bias, variance and mean integrated squared error of the new approach are derived. From the asymptotic mean integrated squared error, a plug-in bandwidth is proposed. Additionally, a bootstrap selector to be used in this context is designed. Through a comprehensive simulation study, the behaviour of the estimator and the bandwidth selectors considering different scenarios of data grouping is shown. The performance of the different approaches is also illustrated with a real grouped emergence data set of Avena sterilis (wild oat).
- Published
- 2019
9. A random effect multiplicative heteroscedastic model for bacterial growth
- Author
-
Quinto Emiliano J, Francisco-Fernández Mario, and Cao Ricardo
- Subjects
Computer applications to medicine. Medical informatics ,R858-859.7 ,Biology (General) ,QH301-705.5 - Abstract
Abstract Background Predictive microbiology develops mathematical models that can predict the growth rate of a microorganism population under a set of environmental conditions. Many primary growth models have been proposed. However, when primary models are applied to bacterial growth curves, the biological variability is reduced to a single curve defined by some kinetic parameters (lag time and growth rate), and sometimes the models give poor fits in some regions of the curve. The development of a prediction band (from a set of bacterial growth curves) using non-parametric and bootstrap methods permits to overcome that problem and include the biological variability of the microorganism into the modelling process. Results Absorbance data from Listeria monocytogenes cultured at 22, 26, 38, and 42°C were selected under different environmental conditions of pH (4.5, 5.5, 6.5, and 7.4) and percentage of NaCl (2.5, 3.5, 4.5, and 5.5). Transformation of absorbance data to viable count data was carried out. A random effect multiplicative heteroscedastic model was considered to explain the dynamics of bacterial growth. The concept of a prediction band for microbial growth is proposed. The bootstrap method was used to obtain resamples from this model. An iterative procedure is proposed to overcome the computer intensive task of calculating simultaneous prediction intervals, along time, for bacterial growth. The bands were narrower below the inflection point (0-8 h at 22°C, and 0-5.5 h at 42°C), and wider to the right of it (from 9 h onwards at 22°C, and from 7 h onwards at 42°C). A wider band was observed at 42°C than at 22°C when the curves reach their upper asymptote. Similar bands have been obtained for 26 and 38°C. Conclusions The combination of nonparametric models and bootstrap techniques results in a good procedure to obtain reliable prediction bands in this context. Moreover, the new iterative algorithm proposed in this paper allows one to achieve exactly the prefixed coverage probability for the prediction band. The microbial growth bands reflect the influence of the different environmental conditions on the microorganism behaviour, helping in the interpretation of the biological meaning of the growth curves obtained experimentally.
- Published
- 2010
- Full Text
- View/download PDF
10. Analysis of interval‐grouped data in weed science: The binnednp Rcpp package.
- Author
-
Barreiro‐Ures, Daniel, Francisco‐Fernández, Mario, Cao, Ricardo, Fraguela, Basilio B., Doallo, Ramón, González‐Andújar, José Luis, and Reyes, Miguel
- Subjects
- *
WEED science , *DATA science , *ENVIRONMENTAL indicators , *DATA analysis , *WILD oat - Abstract
Weed scientists are usually interested in the study of the distribution and density functions of the random variable that relates weed emergence with environmental indices like the hydrothermal time (HTT). However, in many situations, experimental data are presented in a grouped way and, therefore, the standard nonparametric kernel estimators cannot be computed.Kernel estimators for the density and distribution functions for interval‐grouped data, as well as bootstrap confidence bands for these functions, have been proposed and implemented in the binnednp package. Analysis with different treatments can also be performed using a bootstrap approach and a Cramér‐von Mises type distance. Several bandwidth selection procedures were also implemented. This package also allows to estimate different emergence indices that measure the shape of the data distribution. The values of these indices are useful for the selection of the soil depth at which HTT should be measured which, in turn, would maximize the predictive power of the proposed methods.This paper presents the functions of the package and provides an example using an emergence data set of Avena sterilis (wild oat).The binnednp package provides investigators with a unique set of tools allowing the weed science research community to analyze interval‐grouped data. [ABSTRACT FROM AUTHOR]
- Published
- 2019
- Full Text
- View/download PDF
11. Comparing Simultaneous and Pointwise Confidence Intervals for Hydrological Processes.
- Author
-
Francisco-Fernández, Mario and Quintela-del-Río, Alejandro
- Subjects
- *
CONFIDENCE intervals , *HYDROLOGY , *DISTRIBUTION (Probability theory) , *RANDOM variables , *STREAMFLOW - Abstract
Distribution function estimation of the random variable of river flow is an important problem in hydrology. This issue is directly related to quantile estimation, and consequently to return level prediction. The estimation process can be complemented with the construction of confidence intervals (CIs) to perform a probabilistic assessment of the different variables and/or estimated functions. In this work, several methods for constructing CIs using bootstrap techniques, and parametric and nonparametric procedures in the estimation process are studied and compared. In the case that the target is the joint estimation of a vector of values, some new corrections to obtain joint coverage probabilities closer to the corresponding nominal values are also presented. A comprehensive simulation study compares the different approaches, and the application of the different procedures to real data sets from four rivers in the United States and one in Spain complete the paper. [ABSTRACT FROM AUTHOR]
- Published
- 2016
- Full Text
- View/download PDF
12. Nonparametric functional data estimation applied to ozone data: Prediction and extreme value analysis
- Author
-
Quintela-del-Rı´o, Alejandro and Francisco-Fernández, Mario
- Subjects
- *
EXTREME value theory , *AIR pollution , *DATA analysis , *METHODOLOGY , *OZONE layer , *TIME series analysis - Abstract
Abstract: The study of extreme values and prediction of ozone data is an important topic of research when dealing with environmental problems. Classical extreme value theory is usually used in air-pollution studies. It consists in fitting a parametric generalised extreme value (GEV) distribution to a data set of extreme values, and using the estimated distribution to compute return levels and other quantities of interest. Here, we propose to estimate these values using nonparametric functional data methods. Functional data analysis is a relatively new statistical methodology that generally deals with data consisting of curves or multi-dimensional variables. In this paper, we use this technique, jointly with nonparametric curve estimation, to provide alternatives to the usual parametric statistical tools. The nonparametric estimators are applied to real samples of maximum ozone values obtained from several monitoring stations belonging to the Automatic Urban and Rural Network (AURN) in the UK. The results show that nonparametric estimators work satisfactorily, outperforming the behaviour of classical parametric estimators. Functional data analysis is also used to predict stratospheric ozone concentrations. We show an application, using the data set of mean monthly ozone concentrations in Arosa, Switzerland, and the results are compared with those obtained by classical time series (ARIMA) analysis. [ABSTRACT FROM AUTHOR]
- Published
- 2011
- Full Text
- View/download PDF
13. A random effect multiplicative heteroscedastic model for bacterial growth.
- Author
-
Cao, Ricardo, Francisco-Fernández, Mario, and Quinto, Emiliano J
- Subjects
- *
MICROBIOLOGY , *BACTERIAL growth , *MICROORGANISMS , *GRAM-positive bacteria , *FOODBORNE diseases , *MATHEMATICAL models - Abstract
Background: Predictive microbiology develops mathematical models that can predict the growth rate of a microorganism population under a set of environmental conditions. Many primary growth models have been proposed. However, when primary models are applied to bacterial growth curves, the biological variability is reduced to a single curve defined by some kinetic parameters (lag time and growth rate), and sometimes the models give poor fits in some regions of the curve. The development of a prediction band (from a set of bacterial growth curves) using non-parametric and bootstrap methods permits to overcome that problem and include the biological variability of the microorganism into the modelling process. Results: Absorbance data from Listeria monocytogenes cultured at 22, 26, 38, and 42°C were selected under different environmental conditions of pH (4.5, 5.5, 6.5, and 7.4) and percentage of NaCl (2.5, 3.5, 4.5, and 5.5). Transformation of absorbance data to viable count data was carried out. A random effect multiplicative heteroscedastic model was considered to explain the dynamics of bacterial growth. The concept of a prediction band for microbial growth is proposed. The bootstrap method was used to obtain resamples from this model. An iterative procedure is proposed to overcome the computer intensive task of calculating simultaneous prediction intervals, along time, for bacterial growth. The bands were narrower below the inflection point (0-8 h at 22°C, and 0-5.5 h at 42°C), and wider to the right of it (from 9 h onwards at 22°C, and from 7 h onwards at 42°C). A wider band was observed at 42°C than at 22°C when the curves reach their upper asymptote. Similar bands have been obtained for 26 and 38°C. Conclusions: The combination of nonparametric models and bootstrap techniques results in a good procedure to obtain reliable prediction bands in this context. Moreover, the new iterative algorithm proposed in this paper allows one to achieve exactly the prefixed coverage probability for the prediction band. The microbial growth bands reflect the influence of the different environmental conditions on the microorganism behaviour, helping in the interpretation of the biological meaning of the growth curves obtained experimentally. [ABSTRACT FROM AUTHOR]
- Published
- 2010
- Full Text
- View/download PDF
14. Nonparametric Density and Regression Estimation for Samples of Very Large Size
- Author
-
Barreiro-Ures, Daniel, Francisco-Fernández, Mario, and Cao, Ricardo
- Subjects
Redes neuronales (Informática) ,Estadística no paramétrica-Informática - Abstract
Programa Oficial de Doutoramento en Estatística e Investigación Operativa. 555V01 [Abstract] This dissertation mainly deals with the problem of bandwidth selection in the context of nonparametric density and regression estimation for samples of very large size. Some bandwidth selection methods have the disadvantage of high computational complexity. This implies that the number of operations required to compute the bandwidth grows very rapidly as the sample size increases, so that the computational cost associated with these algorithms makes them unsuitable for samples of very large size. In the present thesis, this problem is addressed through the use of subagging, an ensemble method that combines bootstrap aggregating or bagging with the use of subsampling. The latter reduces the computational cost associated with the process of bandwidth selection, while the former is aimed at achieving signi cant reductions in the variability of the bandwidth selector. Thus, subagging versions are proposed for bandwidth selection methods based on widely known criteria such as cross-validation or bootstrap. When applying subagging to the cross-validation bandwidth selector, both for the Parzen{Rosenblatt estimator and the Nadaraya{ Watson estimator, the proposed selectors are studied and their asymptotic properties derived. The empirical behavior of all the proposed bandwidth selectors is shown through various simulation studies and applications to real datasets. [Resumen] Esta disertación aborda principalmente el problema de la selección de la ventana en el contexto de la estimación no paramétrica de la densidad y de la regresión para muestras de gran tamaño. Algunos métodos de selección de la ventana tienen el inconveniente de contar con una elevada complejidad computacional. Esto implica que el número de operaciones necesarias para el cálculo de la ventana crece muy rápidamente a medida que el tamaño muestral aumenta, de manera que el coste computacional asociado a estos algoritmos los hace inadecuados para muestras de gran tamaño. En la presente tesis, este problema se aborda mediante el uso del subagging, un método de aprendizaje conjunto que combina el bootstrap aggregating o bagging con el uso de submuestreo. Este ultimo reduce el coste computacional asociado al proceso de selección de la ventana, mientras que el primero tiene como objetivo conseguir reducciones signi cativas en la variabilidad del selector de la ventana. Así, se proponen versiones subagging para métodos de selección de la ventana basados en criterios ampliamente conocidos, como la validación cruzada o el bootstrap. Al aplicar subagging al selector de la ventana de tipo validación cruzada, tanto para el estimador de Parzen{Rosenblatt como para el estimador de Nadaraya{Watson, se estudian los selectores propuestos y se derivan sus propiedades asintóticas. El comportamiento empírico de todos los selectores de la ventana propuestos se muestra mediante varios estudios de simulación y aplicaciones a conjuntos de datos reales [Resumo] Esta disertación aborda o problema da selección da ventá no contexto da estimación non paramétrica da densidade e da regresión para mostras de gran tamaño. Algúns métodos de selección da ventá teñen o inconveniente de contar cunha alta complexidade computacional. Isto implica que o número de operacións necesarias para o cálculo da ventá crece moi rápidamente a medida que aumenta o tamaño muestral, polo que o coste computacional asociado a estes algoritmos fainos inadecuados para mostras de gran tamaño. Na presente tese, este problema abórdase mediante o uso do subagging, un método de aprendizaxe conxunta que combina o bootstrap aggregating ou bagging co uso de submostraxe. Este último reduce o custo computacional asociado ao proceso de selección da ventá, mentres que o primeiro ten como obxectivo conseguir reducións signi cativas na variabilidade do selector da ventá. Así, propóñense versións subagging para métodos de selección da ventá baseados en criterios amplamente coñecidos, como a validación cruzada ou o bootstrap. Ao aplicar subagging ao selector da ventá de tipo validación cruzada, tanto para o estimador de Parzen{Rosenblatt como para o estimador de Nadaraya{Watson, estúrdanse os selectores propostos e derívanse as súas propiedades asintóticas. O comportamento empírico de todos os selectores da ventá propostos mostrase mediante varios estudos de simulación e aplicacións a conxuntos de datos reais. This research has been supported by MINECO Grant MTM2017-82724-R, and by the Xunta de Galicia (Grupos de Referencia Competitiva ED431C-2016-015, ED431C- 2020-14, Centro Singular de Investigación de Galicia ED431G/01 and Centro de Investigación del Sistema Universitario de Galicia ED431G 2019/01), all of them through the ERDF (European Regional Development Fund). Additionally, this work has been partially carried out during a visit to the Texas A&M University, College Station, financed by INDITEX, with reference INDITEX-UDC 2019. The author is grateful to the Centro de Coordinación de Alertas y Emergencias Sanitarias for kindly providing the COVID-19 hospitalization dataset. Xunta de Galicia; ED431C-2016-015 Xunta de Galicia; ED431C-2020-14 Xunta de Galicia; ED431G/01 Xunta de Galicia; ED431G 2019/01
- Published
- 2021
15. Nonparametric Inference for Regression Models with Spatially Correlated Errors
- Author
-
Meilán-Vila, Andrea, Francisco-Fernández, Mario, and Crujeiras-Casais, Rosa M.
- Subjects
Spatial dependence ,Estimación non paramétrica ,Estimación no paramétrica ,Teoría de la estimación-Modelos matemáticos ,Estadística no paramétrica-Informática ,Estatística circular ,Contraste de bondad de ajuste ,Dependencia espacial ,Goodness-of- t test ,Análisis de regresión ,Regresión lineal-circular ,Estadística circular ,Nonparametric estimation ,Circular statistics ,Linear-circular regression - Abstract
Programa Oficial de Doutoramento en Estatística e Investigación Operativa. 5017V01 [Abstract] Regression estimation can be approached using nonparametric procedures, producing exible estimators and avoiding misspeci cation problems. Alternatively, parametric methods may be preferable to nonparametric approaches if the regression function belongs to the assumed parametric family. However, a bad speci cation of this family can lead to wrong conclusions. Regression function misspeci cation problems can be somewhat tackled by applying a goodness-of- t test. For data presenting some kind of complexity, for example, circular data, the approaches used in regression estimation or in goodness-of- t tests have to be conveniently adapted. Moreover, it might occur that the variables of interest can present a certain type of dependence. For example, they can be spatially correlated, where observations which are close in space tend to be more similar than observations that are far apart. The goal of this thesis is twofold, rst, some inference problems for regression models with Euclidean response and covariates, and spatially correlated errors are analyzed. More speci - cally, a testing procedure for parametric regression models in the presence of spatial correlation is proposed. The second aim is to design and study new approaches to deal with regression function estimation and goodness-of- t tests for models with a circular response and an Rd-valued covariate. In this setting, nonparametric proposals to estimate the circular regression function are provided and studied, under the assumption of independence and also for spatially correlated errors. Moreover, goodness-of- t tests for assessing a parametric regression model are presented in these two frameworks. Comprehensive simulation studies and application of the different techniques to real datasets complete this dissertation. [Resumo] A estimación da regresión pode ser abordada empregando técnicas non paramétricas, dando lugar a estimadores exibles e evitando problemas de mala especi ficación. Alternativamente, os métodos paramétricos poden ser preferibles se a función de regresión pertence á familia paramétrica asumida. Porén, unha mala especi ficación desta familia pode levar a conclusións equivocadas. Os problemas de especi cación incorrecta da función de regresión poden ser abordados aplicando un contraste de bondade de axuste. Para datos que presentan algún tipo de complexidade, por exemplo, datos circulares, os métodos empregados na estimación ou nos contrastes, deben adaptarse convenientemente. Ademais, pode ocorrer que as variables de interese poidan presentar un certo tipo de dependencia. Por exemplo, poden estar espacialmente correladas, onde as observacións que están preto no espazo tenden a ser máis similares que as observacións que están lonxe. O obxectivo desta tese é dobre, primeiro, analízanse problemas de inferencia para modelos de regresión con resposta e covariables Euclídeas, e erros espacialmente correlados. Máis concretamente, contrástase se a función de regresión pertence a unha familia paramétrica, en presenza de correlación espacial. O segundo obxectivo é deseñar e estudar novos procedementos para abordar estimación e contrastes da función regresión para modelos con resposta circular e covariable con valores en Rd. Neste contexto, preséntanse e estúdanse propostas non paramétricas para estimar a función de regresión circular, baixo o suposto de independencia e tamén para erros espacialmente correlados. Ademais, nestes dous contextos, preséntanse contrastes para avaliar un modelo de regresión paramétrico. Esta memoria complétase con estudos de simulación exhaustivos e aplicacións a conxuntos de datos reais. [Resumen] La estimación de la regresión puede ser abordada usando técnicas no paramétricas, dando lugar a estimadores flexibles y evitando problemas de mala especificación. Alternativamente, los métodos paramétricos pueden ser preferibles si la función de regresión pertenece a la familia paramétrica asumida. Sin embargo, una mala especificación de esta familia puede llevar a conclusiones equivocadas. Los problemas de especificación incorrecta de la función de regresión pueden ser abordados aplicando un contraste de bondad de ajuste. Para datos que presentan algún tipo de complejidad, por ejemplo, datos circulares, los métodos utilizados en la estimación o en los contrastes, deben adaptarse convenientemente. Además, puede ocurrir que las variables de interés puedan presentar un cierto tipo de dependencia. Por ejemplo, pueden estar espacialmente correladas, donde las observaciones que están cerca en el espacio tienden a ser más similares que las observaciones que están lejos. El objetivo de esta tesis es doble, primero, se analizan problemas de inferencia para modelos de regresión con respuesta y covariables Euclídeas, y errores espacialmente correlados. Más concretamente, se contrasta si la función de regresión pertenece a una familia paramétrica, en presencia de correlación espacial. El segundo objetivo es diseñar y estudiar nuevos procedimientos para abordar estimación y contrastes de la función regresión para modelos con respuesta circular y covariable con valores en J.Rd. En este contexto, se presentan y estudian propuestas no paramétricas para estimar la función de regresión, bajo el supuesto de independencia y también para errores espacialmente correlados. Además, en estos dos contextos, se presentan contrastes para evaluar un modelo de regresión paramétrico. Esta memoria se completa con estudios de simulación exhaustivos y aplicaciones a conjuntos de datos reales. Palabras clave: contraste de bondad de ajuste, estadística circular, estimación no paramétrica, regresión lineal-circular, dependencia espacial
- Published
- 2020
16. Statistical methods for studying emergence curves in weed science
- Author
-
Reyes Cortés, Miguel Ángel, Cao Abad, Ricardo, Francisco Fernández, Mario, and Universidade da Coruña. Departamento de Matemáticas
- Subjects
Núcleos (Matemáticas) ,Ajuste de curva ,Malas hierbas, Lucha contra-Modelos matemáticos ,Estadística no paramétrica - Abstract
[Resumen] Esta tesis trata el problema de la estimación de la función de densidad y de distribución cuando los datos se presentan agrupados. Para este propósito, se considera el estimador núcleo de la densidad y se propone una modificación para usarlo con datos agrupados. Siempre que Be cumplan los supuestos adecuados, se demuestra que el conocido selector plug-in AMISE óptimo de la ventana puede usarse satisfactoriamente con estos datos, lo que en la práctica lleva a definir el concepto de agrupación ligera. Para escenarios de agrupación pesada, se propone un selector bootstrap. Mediante estudios de simulación se muestra el buen desempeño del estimador cuando se usa adecuadamente el selector plug-in o el selector bootstrap, dependiendo del grado de agrupación de los datos. Con base en el estimador núcleo de la densidad para datos agrupados, Be deriva. un estimador núcleo de la distribución para este tipo de datos. Se obtienen formalmente sus propiedades asintóticas y se estudia su desempeño en diferentes escenarios de agrupación usando un selector plugin adecuado. Finalmente, mediante aplicaciones a datos reales, se muestra la efectividad de los métodos no paramétricos propuestos en esta disertación, mismos que en algunos casos superan el desempeño de algunos métodos paramétricos habitualmente usados en malherbolog!a para C6timar la probabilidad de emergencia de las malas hierbas., [Abstract] This thesis deals with the problem of estimating the density and distributioll functions when the data at hand are grouped. For this, the classical kernel density estimator is considered and a suitable modification is proposed for using it with that type of data. Likewise, whenever the appropriate assumptions are met, it is formally proved that the well-known AMISE optimal plug-in bandwidth selector can be successfully used in the presence of grouped data, which in practice leads to define the concept of light grouping. For scenarios of heavy grouping, an alternative bootstrap bandwidth selector is proposed. By means of simulation studies, it is shown the good performance of the estimator when adequately using either the plug-in or the bootstrap bandwidth selector, depending on the degree of grouping. Based on the kernel density estimator for grouped data, a kernel distribution estimator for grouped data is derived. lis asymptotic properties are formally obtained, and itB performance is studied in different grouping seenarios using a Buitable plug-in selector. Finally, applications to real data coming from weed science show the effectiveness of the nonparametric methods proposed in this dissertation, which in sorne cases outperform the typical parametric methoda used by weed scientists for estimating seedling emergence probabilities., [Resumo] Esta tese trata o problema da estimación da función de densidade e de distribución cando os datos se presentan agrupados. Para este propósito, considérase o estimador núcleo da densidade e proponse unha modificación para usalo con datos agrupados. Sempre que se cumpran os supostos axeitados, demóstrase que o coñecido selector plug-in AMI SE óptimo da ventá pode usarse satisfactoriamente con estes datos, o que na práctica leva a definir o concepto de agrupación lixeira. Para escenarios de agrupación pesada, proponse un selector bootstrap. Mediante estudos de simulación móstrase o bo desempeño do estimador cando Be usa axeitadamente o selector plug-in ou o selector bootstrap, dependendo do grao de agrupación dos datos. Con base no estimador núcleo da densidade para datos agrupados, derivase un estimador núcleo da distribución para este tipo de datos. Obtéñense formalmente as súas propiedades asintóticas e estúdase o seu desempeño en diferentes escenarios de agrupación usando un selector plug-in adecuado. Finalmente, mediante aplicacións a datos reais, móstrase a efectividade dos métodos non paranlétricos propostos nesta disertación, m; mesmos que nalgúns casos superan o desempeño dalgúns métodos paramétricos habitualmente usados en malherboloxía para estimar a probabilidade de emerxencia das malas herbas.
- Published
- 2015
Catalog
Discovery Service for Jio Institute Digital Library
For full access to our library's resources, please sign in.