Impact of OVL Variation on AUC Bias Estimated by Non-parametric Methods

Transmissão através de Videoconferência

Por Carina Silva (Escola Superior de Tecnologia de Saúde de Lisboa, Instituto Politécnico de Lisboa & Centro de Estatística e Aplicações da Universidade de Lisboa).

Abstract: The area under the ROC curve (AUC) is the most commonly used index in the ROC methodology to evaluate the performance of a classifier that discriminates between two mutually exclusive conditions. The AUC can admit values between 0.5 and 1, where values close to 1 indicate that the model of classification has a high discriminative power. The overlap coefficient (OVL) between two density functions is defined as the common area between both functions. This coefficient is used as a measure of agreement between two distributions presenting values between 0 and 1, where values close to 1 reveal total overlapping densities. These two measures were used to construct the arrow plot to select differential expressed genes. A simulation study using the bootstrap method is presented in order to estimate AUC bias and standard error using empirical and kernel methods. In order to assess the impact of the OVL variation on the AUC bias, samples from various distributions were simulated considering different values for its parameters and for fixed OVL values between 0 and 1. Samples of dimensions 15, 30, 50 and 100 and 1000 bootstrap replicates for each scenario were considered.


CEAUL - Centro de Estatística e Aplicações da Universidade de Lisboa / CEMAT - Ciências - Centro de Matemática Computacional e Estocástica

This course will teach how to apply the SWAT eco-hydrological model to assess the impacts of climate and associated changes on water-soil-plant interactions, and consequences to water resources, soil erosion and nutrient cycles.

Mathematical Logic Webinar, por Anton Freund (Technische Universität Darmstadt).

Encontro Scientia, por William Carvalho (Professor Convidado - Faculdade de Ciências Biológicas e Ambientais, Universidade Federal da Grande Dourados, Mato Grosso do Sul).

Geometry Webinar, por Pierre Schapira (Professor emeritus Sorbonne University).

Mapa de Portugal e informações sobre os Censos 2021

Os Censos 2021 estão aí! Esteja atento à sua caixa de correio.

Imagem ilustrativa da ação de formação

Formação de Comunicação em Ciência(s), por Rita Ferreira (InnovPlantProtect, Gestora de Comunicação / NOVA FCSH-ITQB NOVA, Mestranda / Rede SciComPt, sócia).

This course aims to provide students with basic knowledge of R programming, allowing them to manipulate and visualize data with R

This course aims to provide students with statistical knowledge and tools to manipulate, analyze and visualize biological data with R. Introduction to modeling, simulations and Bayesian statistics.

Logótipo da iniciativa

"Património e divulgação em Paleontologia" é o tema escolhido para esta edição do EJIP.

A mostly practical course, offering an overview on different community ecology and macroecological methods and software.

Logótipo da conferência, sobre um fundo branco

THE meeting place for Mobility Management practitioners and experts all over Europe.

An overview of the different ways to measure biodiversity, and provides tips for the stratification of primary biodiversity data and the construction of variables that describe its various facets.

Título e data do evento

An overview of recent developments and applications of the algebraic and analytic theory of D-modules.

This course aims at enabling the participants to use different methods to measure the impacts of pollutants on ecosystems.

How to predict the three-dimensional structure of nucleic acid and proteins from their sequence and use the acquired knowledge to improve or create new methodologies in molecular biology research and diagnostics?

Representação da cidade de Lisboa, acompanhada do título da conferência

The 22nd in the series of triennial conferences which bring together the Particle and Nuclear Physics communities.

Logótipo do evento, sobre um fundo branco

Welcome to an unique silicon for photovoltaics event in Lisbon!

Logótipo da Unite!

Summer School no âmbito da Rede UNITE! (University Network for Innovation, Technology and Engineering), de que faz parte a ULisboa, com inscrições a decorrer até 15 de maio de 2021.