Gilda Soromenho - Academia.edu (original) (raw)

Papers by Gilda Soromenho

Research paper thumbnail of Contributos À Compreensão Da Distribuição De Verbas Destinadas À Educação Nos Municípios Portugueses Em 1870-1880

Interacções, May 20, 2014

No âmbito do projeto Atlas-Repertório dos Municípios na Educação e na Cultura em Portugal (1820-1... more No âmbito do projeto Atlas-Repertório dos Municípios na Educação e na Cultura em Portugal (1820-1986) foram observados diversos indicadores referentes aos anos 1870-1880, em 287 municípios dos 21 distritos do Continente e Ilhas. O objetivo era determinar de que fatores dependia a distribuição de verbas para a Educação a um município nessa época. Nesta perspetiva realizou-se numa primeira análise um estudo exploratório sobre as correlações dos vários indicadores e posteriormente uma análise de regressão múltipla cuja variável dependente é o orçamento geral atribuído ao município. Constatou-se que a distribuição das verbas por município dependia essencialmente de duas variáveis latentes, uma que é o compósito de um conjunto de indicadores sobre a dimensão da população, número de escolas, número de habitantes que só sabem ler ou que sabem ler e escrever e outra que está relacionada com a taxa de alfabetização e com a taxa real de escolarização. Verificou-se também que nesta época já existia uma certa politização na Educação, uma vez que distribuição de verbas dependia da localização do município.

Research paper thumbnail of Perceções de estudantes e professores sobre o ensino, as aprendizagens e a avaliação numa universidade portuguesa

Investigaciones en el contexto universitario actual, 2014, ISBN 978-84-15524-17-5, págs. 535-539, 2014

Research paper thumbnail of Contributos À Compreensão Da Distribuição De Verbas Destinadas À Educação Nos Municípios Portugueses Em 1870-1880

No âmbito do projeto Atlas-Repertorio dos Municipios na Educacao e na Cultura em Portugal (1820-1... more No âmbito do projeto Atlas-Repertorio dos Municipios na Educacao e na Cultura em Portugal (1820-1986) foram observados diversos indicadores referentes aos anos 1870-1880, em 287 municipios dos 21 distritos do Continente e Ilhas. O objetivo era determinar de que fatores dependia a distribuicao de verbas para a Educacao a um municipio nessa epoca. Nesta perspetiva realizou-se numa primeira analise um estudo exploratorio sobre as correlacoes dos varios indicadores e posteriormente uma analise de regressao multipla cuja variavel dependente e o orcamento geral atribuido ao municipio. Constatou-se que a distribuicao das verbas por municipio dependia essencialmente de duas variaveis latentes, uma que e o composito de um conjunto de indicadores sobre a dimensao da populacao, numero de escolas, numero de habitantes que so sabem ler ou que sabem ler e escrever e outra que esta relacionada com a taxa de alfabetizacao e com a taxa real de escolarizacao. Verificou-se tambem que nesta epoca ja ex...

Research paper thumbnail of Comparing approaches for testing the number of components in a finite mixture model

Research paper thumbnail of Comparison of the mixture and the classification maximum likelihood in cluster analysis with binary data

Computational Statistics & Data Analysis, 1996

ABSTRACT

Research paper thumbnail of Misturas de regressões lineares: um novo teste de alteração da estrutura

Research paper thumbnail of Dermatological emergencies: one-year data analysis of 8,620 patients from the largest Portuguese tertiary teaching hospital

European Journal of Dermatology, Sep 1, 2016

Background: Skin conditions frequently lead to emergency department (ED) visits. While most are b... more Background: Skin conditions frequently lead to emergency department (ED) visits. While most are benign in course, some will present as true dermatological urgencies/emergencies, requiring admission. Objectives: To present data on skin diseases most frequently found in the ED, and those most frequently requiring admission at the largest Portuguese tertiary teaching hospital, and to explore an association between epidemiological variables and frequency of diagnoses within this context. Materials and methods: A retrospective study was conducted on all patients examined during 2012 for dermatology emergency consultation (DEC) at the Hospital de Santa Maria, Lisbon, Portugal. Association between epidemiological variables (gender and age of patients, and season of the year) and frequency of diagnoses was investigated. Results: In total, 8,620 patients were examined by a dermatologist in the ED, constituting 3.9% of all ED visits in our centre. Overall, 333 diagnoses were made, the most frequent of which was eczema not-otherwise-specified (9.4%). However, infectious and parasitic diseases constituted the leading motive for DEC (31.5%). Only 264 patients were admitted, with 65 diagnoses leading to admission. Nine diagnoses alone led to 60% of all admissions. Infectious and parasitic diseases constituted the leading cause of admission (34.7%). An association between frequency of diagnoses and gender, age, and season was identified. Conclusion: Despite the variety of dermatological pathologies, only a limited group of diseases was responsible for most of the true dermatological emergencies. Therefore, in the absence of a readily available dermatologist, knowledge of these entities, as well as demographic and environmental data, may help to improve the management of these patients.

Research paper thumbnail of An entropy criterion for assessing the number of clusters in a mixture model

Journal of Classification, Sep 1, 1996

HAL is a multidisciplinary open access archive for the deposit and dissemination of scientific re... more HAL is a multidisciplinary open access archive for the deposit and dissemination of scientific research documents, whether they are published or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. L'archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d'enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

Research paper thumbnail of Testing for structural change in a mixture of linear regressions

Y-nx1 matrix of observations of dependent variable X-nx(k+1) matrix of predictors-(k+1)x1 matrix ... more Y-nx1 matrix of observations of dependent variable X-nx(k+1) matrix of predictors-(k+1)x1 matrix of regression coefficients-random errors; under the assumption of normality,

Research paper thumbnail of Dermatological emergencies: one-year data analysis of 8,620 patients from the largest Portuguese tertiary teaching hospital

European journal of dermatology : EJD, Jan 13, 2016

Skin conditions frequently lead to emergency department (ED) visits. While most are benign in cou... more Skin conditions frequently lead to emergency department (ED) visits. While most are benign in course, some will present as true dermatological urgencies/emergencies, requiring admission. To present data on skin diseases most frequently found in the ED, and those most frequently requiring admission at the largest Portuguese tertiary teaching hospital, and to explore an association between epidemiological variables and frequency of diagnoses within this context. A retrospective study was conducted on all patients examined during 2012 for dermatology emergency consultation (DEC) at the Hospital de Santa Maria, Lisbon, Portugal. Association between epidemiological variables (gender and age of patients, and season of the year) and frequency of diagnoses was investigated. In total, 8,620 patients were examined by a dermatologist in the ED, constituting 3.9% of all ED visits in our centre. Overall, 333 diagnoses were made, the most frequent of which was eczema not-otherwise-specified (9.4%...

Research paper thumbnail of Fitting mixtures of linear regressions

Journal of Statistical Computation and Simulation, 2010

In most applications, the parameters of a mixture of linear regression models are estimated by ma... more In most applications, the parameters of a mixture of linear regression models are estimated by maximum likelihood using the expectation maximization (EM) algorithm. In this article, we propose the comparison of three algorithms to compute maximum likelihood estimates of the parameters of these models: the EM algorithm, the classification EM algorithm and the stochastic EM algorithm. The comparison of the three procedures was done through a simulation study of the performance (computational effort, statistical properties of estimators and goodness of fit) of these approaches on simulated data sets. Simulation results show that the choice of the approach depends essentially on the configuration of the true regression lines and the initialization of the algorithms.

Research paper thumbnail of An entropy criterion for assessing the number of clusters in a mixture model

Journal of Classification, 1996

HAL is a multidisciplinary open access archive for the deposit and dissemination of scientific re... more HAL is a multidisciplinary open access archive for the deposit and dissemination of scientific research documents, whether they are published or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. L'archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d'enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

Research paper thumbnail of Fitting mixtures of linear regressions

Journal of Statistical Computation and Simulation, Feb 1, 2010

In most applications, the parameters of a mixture of linear regression models are estimated by ma... more In most applications, the parameters of a mixture of linear regression models are estimated by maximum likelihood using the expectation maximization (EM) algorithm. In this article, we propose the comparison of three algorithms to compute maximum likelihood estimates of the parameters of these models: the EM algorithm, the classification EM algorithm and the stochastic EM algorithm. The comparison of the three procedures was done through a simulation study of the performance (computational effort, statistical properties of estimators and goodness of fit) of these approaches on simulated data sets. Simulation results show that the choice of the approach depends essentially on the configuration of the true regression lines and the initialization of the algorithms.

Research paper thumbnail of Comparison of EM and SEM Algorithms in Poisson Regression Models: A Simulation Study

Communications in Statistics - Simulation and Computation, Dec 20, 2011

In this work, we propose to compare two algorithms to compute maximum likelihood estimates for th... more In this work, we propose to compare two algorithms to compute maximum likelihood estimates for the parameters of a mixture Poisson regression model: the EM algorithm and the Stochastic EM algorithm. The comparison of the two procedures was done through a simulation study of the performance of these approaches on simulated data sets and real data sets. Simulation results show that the choice of the approach depends essentially on the overlap of the regression lines. In the real data case, we show that the Stochastic EM algorithm resulted in model estimates that best fit the regression model.

Research paper thumbnail of Misturas de regressões lineares: um novo teste de alteração da estrutura

Research paper thumbnail of Measuring the component overlapping in mixtures of linear regressions

Entropy-type measures for the heterogeneity of data have been used for a long time. In a mixture ... more Entropy-type measures for the heterogeneity of data have been used for a long time. In a mixture model context, entropy criterions can be used to measure the overlapping of the mixture components. In this paper we study an entropy-based criterion in mixtures of linear regressions to measure the closeness between the mixture components. We show how an entropy criterion can be derived based on the Kullback-Leiber distance, which is a measure of distance between probability distributions. To investigate the effectiveness of the proposed criterion, a simulation study was performed.

Research paper thumbnail of Comparison of Mixture and Classification Maximum Likelihood Approaches in Poisson Regression Models

In this work, we propose to compare two algorithms to compute maximum likelihood estimators of th... more In this work, we propose to compare two algorithms to compute maximum likelihood estimators of the parameters of a mixture Poisson regression models. To estimate these parameters, we may use the EM algorithm in a mixture approach or the CEM algorithm in a classification approach. The comparison of the two procedures was done through a simulation study of the performance of these approaches on simulated data sets in a target number of iterations. Simulation results show that the CEM algorithm is a good alternative to the EM algorithm for fitting Poisson mixture regression models, having the advantage of converging more quickly.

Research paper thumbnail of Joclad 2006

Neste trabalho, descrevemos um teste de alteração da estrutura que desenvolvemos para o caso de m... more Neste trabalho, descrevemos um teste de alteração da estrutura que desenvolvemos para o caso de modelos de mistura de regressões lineares. Este teste baseia-se na comparação entre o modelo de mistura estimado a partir do conjunto de observações iniciais e o modelo de mistura estimado a partir da totalidade das observações disponíveis (observações iniciais e bservações novas). Com o objectivo de ilustrar a aplicação deste teste em situações práticas onde as misturas de regressões lineares são adequadas, apresentam-se alguns exemplos de aplicação recorrendo a amostras geradas.Universidade Lusíada de Lisboa, EMBI - CLEG Secção de Estudos em Marketing e Business Intelligence do Centro Lusíada de Economia e Gestão , Centro de Estatística e Aplicações – CEAUL, Société Francophone de Classification, German Classification Society , APODEMO - Associação Portuguesa de Empresas de Estudos de Mercado e Opinião, PSE - Soluções analíticas para suporte à decisão / SPSS , SAS Portugal , DataScience...

Research paper thumbnail of Internet risk behaviours: a study developed in a secondary school

El uso seguro de Internet ha atraído la atención de los medios comunicación social, que realzan f... more El uso seguro de Internet ha atraído la atención de los medios comunicación social, que realzan fundamentalmente los riesgos para los jóvenes. También nos encontramos en las escuelas con algunos casos problemáticos relacionados con la utilización incorrecta de los ordenadores y de internet, en sus vertientes técnicas y éticas y aún ciertos comportamientos de riesgo que pueden tener consecuencias nefastas. Consideramos importante profundizar en este problema y estudiar su dimensión en una escuela de enseñanza secundaria situada en la región de Lisboa para delinear un proyecto de intervención. Después de una revisión de la literatura, optamos una investigación con cuestionario, haciendo una adaptación de tres versiones: una para los jóvenes, otro para los padres y otro para los profesores. Los resultados que hemos encontrado en esta fase inicial del estudio indican que algunos estudiantes tienen comportamientos de riesgo, tales como el comportamiento agresivo, la publicación de la inf...

Research paper thumbnail of Questionário de Conhecimento sobre Literatura para a Infância

A investigação na área da Educação realça como o livro de literatura para a infância pode contrib... more A investigação na área da Educação realça como o livro de literatura para a infância pode contribuir para o desenvolvimento e aprendizagem das crianças. No âmbito da Educação Pré-escolar, esta realidade sugere a necessidade de avaliar a perceção dos educadores sobre o livro de literatura para a infância. O presente artigo apresenta o processo de construção e validação de um Questionário de Conhecimento sobre a Literatura para a Infância (QCLI), pretendendo assim contribuir com uma resposta a essa necessidade. Participaram no estudo 270 educadores de infância em Portugal continental, dos quais 99% dos inquiridos são do sexo feminino com uma média de idades de 43 anos. As respostas ao questionário, após diversos procedimentos de validação, conduziram pela análise fatorial confirmatória composta a três dimensões: uma primeira sobre a perceção dos educadores sobre a competência pessoal no uso do livro de literatura para a infância, uma segunda sobre a perceção dos educadores acerca da p...

Research paper thumbnail of Contributos À Compreensão Da Distribuição De Verbas Destinadas À Educação Nos Municípios Portugueses Em 1870-1880

Interacções, May 20, 2014

No âmbito do projeto Atlas-Repertório dos Municípios na Educação e na Cultura em Portugal (1820-1... more No âmbito do projeto Atlas-Repertório dos Municípios na Educação e na Cultura em Portugal (1820-1986) foram observados diversos indicadores referentes aos anos 1870-1880, em 287 municípios dos 21 distritos do Continente e Ilhas. O objetivo era determinar de que fatores dependia a distribuição de verbas para a Educação a um município nessa época. Nesta perspetiva realizou-se numa primeira análise um estudo exploratório sobre as correlações dos vários indicadores e posteriormente uma análise de regressão múltipla cuja variável dependente é o orçamento geral atribuído ao município. Constatou-se que a distribuição das verbas por município dependia essencialmente de duas variáveis latentes, uma que é o compósito de um conjunto de indicadores sobre a dimensão da população, número de escolas, número de habitantes que só sabem ler ou que sabem ler e escrever e outra que está relacionada com a taxa de alfabetização e com a taxa real de escolarização. Verificou-se também que nesta época já existia uma certa politização na Educação, uma vez que distribuição de verbas dependia da localização do município.

Research paper thumbnail of Perceções de estudantes e professores sobre o ensino, as aprendizagens e a avaliação numa universidade portuguesa

Investigaciones en el contexto universitario actual, 2014, ISBN 978-84-15524-17-5, págs. 535-539, 2014

Research paper thumbnail of Contributos À Compreensão Da Distribuição De Verbas Destinadas À Educação Nos Municípios Portugueses Em 1870-1880

No âmbito do projeto Atlas-Repertorio dos Municipios na Educacao e na Cultura em Portugal (1820-1... more No âmbito do projeto Atlas-Repertorio dos Municipios na Educacao e na Cultura em Portugal (1820-1986) foram observados diversos indicadores referentes aos anos 1870-1880, em 287 municipios dos 21 distritos do Continente e Ilhas. O objetivo era determinar de que fatores dependia a distribuicao de verbas para a Educacao a um municipio nessa epoca. Nesta perspetiva realizou-se numa primeira analise um estudo exploratorio sobre as correlacoes dos varios indicadores e posteriormente uma analise de regressao multipla cuja variavel dependente e o orcamento geral atribuido ao municipio. Constatou-se que a distribuicao das verbas por municipio dependia essencialmente de duas variaveis latentes, uma que e o composito de um conjunto de indicadores sobre a dimensao da populacao, numero de escolas, numero de habitantes que so sabem ler ou que sabem ler e escrever e outra que esta relacionada com a taxa de alfabetizacao e com a taxa real de escolarizacao. Verificou-se tambem que nesta epoca ja ex...

Research paper thumbnail of Comparing approaches for testing the number of components in a finite mixture model

Research paper thumbnail of Comparison of the mixture and the classification maximum likelihood in cluster analysis with binary data

Computational Statistics & Data Analysis, 1996

ABSTRACT

Research paper thumbnail of Misturas de regressões lineares: um novo teste de alteração da estrutura

Research paper thumbnail of Dermatological emergencies: one-year data analysis of 8,620 patients from the largest Portuguese tertiary teaching hospital

European Journal of Dermatology, Sep 1, 2016

Background: Skin conditions frequently lead to emergency department (ED) visits. While most are b... more Background: Skin conditions frequently lead to emergency department (ED) visits. While most are benign in course, some will present as true dermatological urgencies/emergencies, requiring admission. Objectives: To present data on skin diseases most frequently found in the ED, and those most frequently requiring admission at the largest Portuguese tertiary teaching hospital, and to explore an association between epidemiological variables and frequency of diagnoses within this context. Materials and methods: A retrospective study was conducted on all patients examined during 2012 for dermatology emergency consultation (DEC) at the Hospital de Santa Maria, Lisbon, Portugal. Association between epidemiological variables (gender and age of patients, and season of the year) and frequency of diagnoses was investigated. Results: In total, 8,620 patients were examined by a dermatologist in the ED, constituting 3.9% of all ED visits in our centre. Overall, 333 diagnoses were made, the most frequent of which was eczema not-otherwise-specified (9.4%). However, infectious and parasitic diseases constituted the leading motive for DEC (31.5%). Only 264 patients were admitted, with 65 diagnoses leading to admission. Nine diagnoses alone led to 60% of all admissions. Infectious and parasitic diseases constituted the leading cause of admission (34.7%). An association between frequency of diagnoses and gender, age, and season was identified. Conclusion: Despite the variety of dermatological pathologies, only a limited group of diseases was responsible for most of the true dermatological emergencies. Therefore, in the absence of a readily available dermatologist, knowledge of these entities, as well as demographic and environmental data, may help to improve the management of these patients.

Research paper thumbnail of An entropy criterion for assessing the number of clusters in a mixture model

Journal of Classification, Sep 1, 1996

HAL is a multidisciplinary open access archive for the deposit and dissemination of scientific re... more HAL is a multidisciplinary open access archive for the deposit and dissemination of scientific research documents, whether they are published or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. L'archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d'enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

Research paper thumbnail of Testing for structural change in a mixture of linear regressions

Y-nx1 matrix of observations of dependent variable X-nx(k+1) matrix of predictors-(k+1)x1 matrix ... more Y-nx1 matrix of observations of dependent variable X-nx(k+1) matrix of predictors-(k+1)x1 matrix of regression coefficients-random errors; under the assumption of normality,

Research paper thumbnail of Dermatological emergencies: one-year data analysis of 8,620 patients from the largest Portuguese tertiary teaching hospital

European journal of dermatology : EJD, Jan 13, 2016

Skin conditions frequently lead to emergency department (ED) visits. While most are benign in cou... more Skin conditions frequently lead to emergency department (ED) visits. While most are benign in course, some will present as true dermatological urgencies/emergencies, requiring admission. To present data on skin diseases most frequently found in the ED, and those most frequently requiring admission at the largest Portuguese tertiary teaching hospital, and to explore an association between epidemiological variables and frequency of diagnoses within this context. A retrospective study was conducted on all patients examined during 2012 for dermatology emergency consultation (DEC) at the Hospital de Santa Maria, Lisbon, Portugal. Association between epidemiological variables (gender and age of patients, and season of the year) and frequency of diagnoses was investigated. In total, 8,620 patients were examined by a dermatologist in the ED, constituting 3.9% of all ED visits in our centre. Overall, 333 diagnoses were made, the most frequent of which was eczema not-otherwise-specified (9.4%...

Research paper thumbnail of Fitting mixtures of linear regressions

Journal of Statistical Computation and Simulation, 2010

In most applications, the parameters of a mixture of linear regression models are estimated by ma... more In most applications, the parameters of a mixture of linear regression models are estimated by maximum likelihood using the expectation maximization (EM) algorithm. In this article, we propose the comparison of three algorithms to compute maximum likelihood estimates of the parameters of these models: the EM algorithm, the classification EM algorithm and the stochastic EM algorithm. The comparison of the three procedures was done through a simulation study of the performance (computational effort, statistical properties of estimators and goodness of fit) of these approaches on simulated data sets. Simulation results show that the choice of the approach depends essentially on the configuration of the true regression lines and the initialization of the algorithms.

Research paper thumbnail of An entropy criterion for assessing the number of clusters in a mixture model

Journal of Classification, 1996

HAL is a multidisciplinary open access archive for the deposit and dissemination of scientific re... more HAL is a multidisciplinary open access archive for the deposit and dissemination of scientific research documents, whether they are published or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. L'archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d'enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

Research paper thumbnail of Fitting mixtures of linear regressions

Journal of Statistical Computation and Simulation, Feb 1, 2010

In most applications, the parameters of a mixture of linear regression models are estimated by ma... more In most applications, the parameters of a mixture of linear regression models are estimated by maximum likelihood using the expectation maximization (EM) algorithm. In this article, we propose the comparison of three algorithms to compute maximum likelihood estimates of the parameters of these models: the EM algorithm, the classification EM algorithm and the stochastic EM algorithm. The comparison of the three procedures was done through a simulation study of the performance (computational effort, statistical properties of estimators and goodness of fit) of these approaches on simulated data sets. Simulation results show that the choice of the approach depends essentially on the configuration of the true regression lines and the initialization of the algorithms.

Research paper thumbnail of Comparison of EM and SEM Algorithms in Poisson Regression Models: A Simulation Study

Communications in Statistics - Simulation and Computation, Dec 20, 2011

In this work, we propose to compare two algorithms to compute maximum likelihood estimates for th... more In this work, we propose to compare two algorithms to compute maximum likelihood estimates for the parameters of a mixture Poisson regression model: the EM algorithm and the Stochastic EM algorithm. The comparison of the two procedures was done through a simulation study of the performance of these approaches on simulated data sets and real data sets. Simulation results show that the choice of the approach depends essentially on the overlap of the regression lines. In the real data case, we show that the Stochastic EM algorithm resulted in model estimates that best fit the regression model.

Research paper thumbnail of Misturas de regressões lineares: um novo teste de alteração da estrutura

Research paper thumbnail of Measuring the component overlapping in mixtures of linear regressions

Entropy-type measures for the heterogeneity of data have been used for a long time. In a mixture ... more Entropy-type measures for the heterogeneity of data have been used for a long time. In a mixture model context, entropy criterions can be used to measure the overlapping of the mixture components. In this paper we study an entropy-based criterion in mixtures of linear regressions to measure the closeness between the mixture components. We show how an entropy criterion can be derived based on the Kullback-Leiber distance, which is a measure of distance between probability distributions. To investigate the effectiveness of the proposed criterion, a simulation study was performed.

Research paper thumbnail of Comparison of Mixture and Classification Maximum Likelihood Approaches in Poisson Regression Models

In this work, we propose to compare two algorithms to compute maximum likelihood estimators of th... more In this work, we propose to compare two algorithms to compute maximum likelihood estimators of the parameters of a mixture Poisson regression models. To estimate these parameters, we may use the EM algorithm in a mixture approach or the CEM algorithm in a classification approach. The comparison of the two procedures was done through a simulation study of the performance of these approaches on simulated data sets in a target number of iterations. Simulation results show that the CEM algorithm is a good alternative to the EM algorithm for fitting Poisson mixture regression models, having the advantage of converging more quickly.

Research paper thumbnail of Joclad 2006

Neste trabalho, descrevemos um teste de alteração da estrutura que desenvolvemos para o caso de m... more Neste trabalho, descrevemos um teste de alteração da estrutura que desenvolvemos para o caso de modelos de mistura de regressões lineares. Este teste baseia-se na comparação entre o modelo de mistura estimado a partir do conjunto de observações iniciais e o modelo de mistura estimado a partir da totalidade das observações disponíveis (observações iniciais e bservações novas). Com o objectivo de ilustrar a aplicação deste teste em situações práticas onde as misturas de regressões lineares são adequadas, apresentam-se alguns exemplos de aplicação recorrendo a amostras geradas.Universidade Lusíada de Lisboa, EMBI - CLEG Secção de Estudos em Marketing e Business Intelligence do Centro Lusíada de Economia e Gestão , Centro de Estatística e Aplicações – CEAUL, Société Francophone de Classification, German Classification Society , APODEMO - Associação Portuguesa de Empresas de Estudos de Mercado e Opinião, PSE - Soluções analíticas para suporte à decisão / SPSS , SAS Portugal , DataScience...

Research paper thumbnail of Internet risk behaviours: a study developed in a secondary school

El uso seguro de Internet ha atraído la atención de los medios comunicación social, que realzan f... more El uso seguro de Internet ha atraído la atención de los medios comunicación social, que realzan fundamentalmente los riesgos para los jóvenes. También nos encontramos en las escuelas con algunos casos problemáticos relacionados con la utilización incorrecta de los ordenadores y de internet, en sus vertientes técnicas y éticas y aún ciertos comportamientos de riesgo que pueden tener consecuencias nefastas. Consideramos importante profundizar en este problema y estudiar su dimensión en una escuela de enseñanza secundaria situada en la región de Lisboa para delinear un proyecto de intervención. Después de una revisión de la literatura, optamos una investigación con cuestionario, haciendo una adaptación de tres versiones: una para los jóvenes, otro para los padres y otro para los profesores. Los resultados que hemos encontrado en esta fase inicial del estudio indican que algunos estudiantes tienen comportamientos de riesgo, tales como el comportamiento agresivo, la publicación de la inf...

Research paper thumbnail of Questionário de Conhecimento sobre Literatura para a Infância

A investigação na área da Educação realça como o livro de literatura para a infância pode contrib... more A investigação na área da Educação realça como o livro de literatura para a infância pode contribuir para o desenvolvimento e aprendizagem das crianças. No âmbito da Educação Pré-escolar, esta realidade sugere a necessidade de avaliar a perceção dos educadores sobre o livro de literatura para a infância. O presente artigo apresenta o processo de construção e validação de um Questionário de Conhecimento sobre a Literatura para a Infância (QCLI), pretendendo assim contribuir com uma resposta a essa necessidade. Participaram no estudo 270 educadores de infância em Portugal continental, dos quais 99% dos inquiridos são do sexo feminino com uma média de idades de 43 anos. As respostas ao questionário, após diversos procedimentos de validação, conduziram pela análise fatorial confirmatória composta a três dimensões: uma primeira sobre a perceção dos educadores sobre a competência pessoal no uso do livro de literatura para a infância, uma segunda sobre a perceção dos educadores acerca da p...