Modelagem univariada e regressão múltipla na análise da oferta e do preço do boi gordo no mercado brasileiro

Dissertação (mestrado)—Universidade de Brasília, Faculdade de Agronomia e Medicina Veterinária, Programa de Pós-Graduação em Agronegócios, 2016.

Análise dos indicadores de qualidade versus taxa de abandono utilizando método de regressão múltipla para serviço de banda larga

Telecommunication is one of the most dynamic and strategic areas in the world. Many technological innovations has modified the way information is exchanged. Information and knowledge are now shared in networks. Broadband Internet is the new way of sharing contents and information. This dissertation deals with performance indicators related to maintenance services of telecommunications networks and uses models of multivariate regression to estimate churn, which is the loss of customers to other companies. In a competitive environment, telecommunications companies have devised strategies to minimize the loss of customers. Loosing customers presents a higher cost than obtaining new ones. Corporations have plenty of data stored in a diversity of databases. Usually the data are not explored properly. This work uses the Knowledge Discovery in Databases (KDD) to establish rules and new models to explain how churn, as a dependent variable, are related to a diversity of service indicators, such as time to deploy the service (in hours), time to repair (in hours), and so on. Extraction of meaningful knowledge is, in many cases, a challenge. Models were tested and statistically analyzed. The work also shows results that allows the analysis and identification of which quality services indicators influence the churn. Actions are also proposed to solve, at least in part, this problem

Atributos de solo determinantes para a estimativa do índice de estabilidade de agregados

O objetivo deste trabalho foi identificar atributos de solo importantes para a estimativa do índice de estabilidade de agregados. Para a identificação dos atributos, utilizou-se a regressão múltipla "stepwise" em dados coletados de Latossolo Vermelho eutroférrico, cultivado em sistema de plantio direto. Para a obtenção dos modelos de estimativa do índice de estabilidade de agregados (IeaE), foram analisadas dez variáveis independentes: argila total, densidade do solo, resistência à penetração, matéria orgânica, pH em água, capacidade efetiva de troca de cátions, água disponível, volume total de poros, macroporosidade e umidade gravimétrica. Os atributos argila total (Arg), densidade do solo (Ds), resistência à penetração (RP), matéria orgânica (MO) e pH são os mais importantes para a estimativa do IeaE, e a função de pedotrânsferência que melhor descreve esse índice é: IeaE = 27,80 + 0,04Arg + 27,97Ds + 12,35RP + 0,45MO - 9,72pH (R² = 0,89).

Modelagem de atributos físico-hídricos do solo numa bacia hidrográfica da região do Alto Rio Grande, MG

O presente trabalho objetivou desenvolver modelos para estimativa dos atributos físico-hídricos do solo porosidade drenável (PD) e capacidade total de retenção de água (CTA), com base em atributos físicos de fácil e rápida obtenção (textura, argila dispersa em água, volume total de poros, densidade do solo e densidade de partícula) e a determinação de matéria orgânica. Os atributos foram coletados na bacia hidrográfica do Ribeirão Marcela, na camada de 0 a 15 cm, obedecendo aos grids de 240 x 240 e 60 x 60 m, totalizando 165 pontos amostrados. Trabalhou-se com regressão múltipla linear, constituindo-se variáveis por meio da combinação dos diferentes atributos entre si e estimando os respectivos coeficientes pelo método dos mínimos quadrados, utilizando-se o programa SAS for Windows. Os modelos gerados para CTA e PD apresentaram boa qualidade estatística, com elevados coeficientes de determinação, baixos erros de estimativa e variáveis significativas, o que permitiu sua aplicação na estimativa desses atributos na bacia hidrográfica do Ribeirão Marcela, representativa do domínio dos Latossolos na região do Alto Rio Grande, fornecendo assim subsídios para implantação de projetos agrícolas e ambientais, especialmente no auxílio à parametrização de modelos de simulação hidrossedimentológica, os quais estão em desenvolvimento para aplicação na região.

Modelagem da Densidade do Solo em Profundidade sob Vegetação Nativa em Minas Gerais

A densidade do solo (Ds) é um importante indicador da qualidade física do solo, mas há pouca disponibilidade de informações sobre seus valores a maiores profundidades em razão da dificuldade amostral envolvida. Portanto, funções de pedotransferência têm sido utilizadas para estimar a Ds com relativo êxito, mas ainda sem especificidade aos diferentes biomas brasileiros. O objetivo deste trabalho foi desenvolver funções matemáticas capazes de descrever a Ds até 1 m de profundidade em áreas de vegetação nativa das regiões central e sul de Minas Gerais. A Ds foi amostrada pelo método do anel volumétrico em 53 perfis de solo de diferentes ordens, em seis profundidades (0-5, 5-10, 10-20, 30-40, 50-60 e 90-100 cm). A Ds variou entre 0,66 e 1,74 kg dm-3, com média de 1,25 kg dm-3, e foi geralmente menor nas camadas de 0-5 e 5-10 cm. Por meio de regressão linear múltipla (stepwise), foram gerados modelos com base nas propriedades químicas de rotina e granulometria, que permitiram estimar a Ds até 1 m de profundidade. Os teores de C orgânico do solo, areia, silte e argila e a capacidade de troca catiônica potencial (T) foram as variáveis de maior relevância nos modelos, que alcançaram maior acurácia para a ordem Latossolos (R2ajust = 0,85), seguida por Cambissolos (R2ajust = 0,69), Nitossolos (R2ajust = 0,67) e Argissolos (R2ajust = 0,51). Uma vez que a modelagem para a base de dados completa atingiu R2ajust de 0,50, pode-se concluir que a estratificação por ordem taxonômica foi útil para melhorar os ajustes obtidos, com exceção da ordem Argissolo.

Modelagem geoestatística por geoprocessamento em uma área da Baixada Santista

Pós-graduação em Geociências e Meio Ambiente - IGCE

Modelagem e otimização por metodologia de superfícies de resposta: um estudo em arames de aço SAE 9254 para molas automobilísticas

Pós-graduação em Engenharia Mecânica - FEG

Previsão de demanda de veículos automotores através do modelo de regressão linear múltipla

This study aimed to model a equation for the demand of automobiles and light commercial vehicles, based on the data from February 2007 to July 2014, through a multiple regression analysis. The literature review consists of an information collection of the history of automotive industry, and it has contributed to the understanding of the current crisis that affects this market, which consequence was a large reduction in sales. The model developed was evaluated by a residual analysis and also was used an adhesion test - F test - with a significance level of 5%. In addition, a coefficient of determination (R2) of 0.8159 was determined, indicating that 81.59% of the demand for automobiles and light commercial vehicles can be explained by the regression variables: interest rate, unemployment rate, broad consumer price index (CPI), gross domestic product (GDP) and tax on industrialized products (IPI). Finally, other ten samples, from August 2014 to May 2015, were tested in the model in order to validate its forecasting quality. Finally, a Monte Carlo Simulation was run in order to obtain a distribution of probabilities of future demands. It was observed that the actual demand in the period after the sample was in the range that was most likely to occur, and that the GDP and the CPI are the variable that have the greatest influence on the developed model

Previsão de demanda de veículos automotores através do modelo de regressão linear múltipla

This study aimed to model a equation for the demand of automobiles and light commercial vehicles, based on the data from February 2007 to July 2014, through a multiple regression analysis. The literature review consists of an information collection of the history of automotive industry, and it has contributed to the understanding of the current crisis that affects this market, which consequence was a large reduction in sales. The model developed was evaluated by a residual analysis and also was used an adhesion test - F test - with a significance level of 5%. In addition, a coefficient of determination (R2) of 0.8159 was determined, indicating that 81.59% of the demand for automobiles and light commercial vehicles can be explained by the regression variables: interest rate, unemployment rate, broad consumer price index (CPI), gross domestic product (GDP) and tax on industrialized products (IPI). Finally, other ten samples, from August 2014 to May 2015, were tested in the model in order to validate its forecasting quality. Finally, a Monte Carlo Simulation was run in order to obtain a distribution of probabilities of future demands. It was observed that the actual demand in the period after the sample was in the range that was most likely to occur, and that the GDP and the CPI are the variable that have the greatest influence on the developed model

Aplicação do escore MELD em pacientes submetidos a transplante de fígado: análise retrospectiva da sobrevida e dos fatores preditivos a curto e longo prazo

BACKGROUND: The model for end-stage liver disease (MELD) was developed to predict short-term mortality in patients with cirrhosis. There are few reports studying the correlation between MELD and long-term posttransplantation survival. AIM: To assess the value of pretransplant MELD in the prediction of posttransplant survival. METHODS: The adult patients (age >18 years) who underwent liver transplantation were examined in a retrospective longitudinal cohort of patients, through the prospective data base. We excluded acute liver failure, retransplantation and reduced or split-livers. The liver donors were evaluated according to: age, sex, weight, creatinine, bilirubin, sodium, aspartate aminotransferase, personal antecedents, brain death cause, steatosis, expanded criteria donor number and index donor risk. The recipients' data were: sex, age, weight, chronic hepatic disease, Child-Turcotte-Pugh points, pretransplant and initial MELD score, pretransplant creatinine clearance, sodium, cold and warm ischemia times, hospital length of stay, blood requirements, and alanine aminotransferase (ALT >1,000 UI/L = liver dysfunction). The Kaplan-Meier method with the log-rank test was used for the univariable analyses of posttransplant patient survival. For the multivariable analyses the Cox proportional hazard regression method with the stepwise procedure was used with stratifying sodium and MELD as variables. ROC curve was used to define area under the curve for MELD and Child-Turcotte-Pugh. RESULTS: A total of 232 patients with 10 years follow up were available. The MELD cutoff was 20 and Child-Turcotte-Pugh cutoff was 11.5. For MELD score > 20, the risk factors for death were: red cell requirements, liver dysfunction and donor's sodium. For the patients with hyponatremia the risk factors were: negative delta-MELD score, red cell requirements, liver dysfunction and donor's sodium. The regression univariated analyses came up with the following risk factors for death: score MELD > 25, blood requirements, recipient creatinine clearance pretransplant and age donor >50. After stepwise analyses, only red cell requirement was predictive. Patients with MELD score < 25 had a 68.86%, 50,44% and 41,50% chance for 1, 5 and 10-year survival and > 25 were 39.13%, 29.81% and 22.36% respectively. Patients without hyponatremia were 65.16%, 50.28% and 41,98% and with hyponatremia 44.44%, 34.28% and 28.57% respectively. Patients with IDR > 1.7 showed 53.7%, 27.71% and 13.85% and index donor risk <1.7 was 63.62%, 51.4% and 44.08%, respectively. Age donor > 50 years showed 38.4%, 26.21% and 13.1% and age donor <50 years showed 65.58%, 26.21% and 13.1%. Association with delta-MELD score did not show any significant difference. Expanded criteria donors were associated with primary non-function and severe liver dysfunction. Predictive factors for death were blood requirements, hyponatremia, liver dysfunction and donor's sodium. CONCLUSION: In conclusion MELD over 25, recipient's hyponatremia, blood requirements, donor's sodium were associated with poor survival.

Avaliação da satisfação dos utentes dos serviços farmacêuticos da Unidade Local de Saúde do Norte Alentejano

Mestrado em Intervenção Sócio-Organizacional na Saúde - Área de especialização: Políticas de Administração e Gestão de Serviços de Saúde.

Determinants of mother-baby relationship evaluated during pregnancy

ABSTRACT Background Mental health promotion is supported by a strong body of knowledge and is a matter of public health with the potential of a large impact on society. Mental health promotion programs should be implemented as soon as possible in life, preferably starting during pregnancy. Programs should focus on malleable determinants, introducing strategies to reduce risk factors or their impact on mother and child, and also on strengthening protective factors to increase resilience. The ambition of early detecting risk situations requires the development and use of tools to assess risk, and the creation of a responsive network of services based in primary health care, especially maternal consultation during pregnancy and the first months of the born child. The number of risk factors and the way they interact and are buffered by protective factors are relevant for the final impact. Maternal-fetal attachment (MFA) is not yet a totally understood and well operationalized concept. Methodological problems limit the comparison of data as many studies used small size samples, had an exploratory character or used different selection criteria and different measures. There is still a lack of studies in high risk populations evaluating the consequences of a weak MFA. Instead, the available studies are not very conclusive, but suggest that social support, anxiety and depression, self-esteem and self-control and sense of coherence are correlated with MFA. MFA is also correlated with health practices during pregnancy, that influence pregnancy and baby outcomes. MFA seems a relevant concept for the future mother baby interaction, but more studies are needed to clarify the concept and its operationalization. Attachment is a strong scientific concept with multiple implications for future child development, personality and relationship with others. Secure attachment is considered an essential basis of good mental health, and promoting mother-baby interaction offers an excellent opportunity to intervention programmes targeted at enhancing mental health and well-being. Understanding the process of attachment and intervening to improve attachment requires a comprehension of more proximal factors, but also a broader approach that assesses the impact of more distal social conditions on attachment and how this social impact is mediated by family functioning and mother-baby interaction. Finally, it is essential to understand how this knowledge could be translated in effective mental health promoting interventions and measures that could reach large populations of pregnant mothers and families. Strengthening emotional availability (EA) seems to be a relevant approach to improve the mother-baby relationship. In this review we have offered evidence suggesting a range of determinants of mother-infant relationship, including age, marital relationship, social disadvantages, migration, parental psychiatric disorders and the situations of abuse or neglect. Based on this theoretical background we constructed a theoretical model that included proximal and distal factors, risk and protective factors, including variables related to the mother, the father, their social support and mother baby interaction from early pregnancy until six months after birth. We selected the Antenatal Psychosocial Health Assessment (ALPHA) for use as an instrument to detect psychosocial risk during pregnancy. Method Ninety two pregnant women were recruited from the Maternal Health Consultation in Primary Health Care (PHC) at Amadora. They had three moments of assessment: at T1 (until 12 weeks of pregnancy) they filed out a questionnaire that included socio-demographic data, ALPHA, Edinburgh post-natal Depression Scale (EDPS), General Health Questionnaire (GHQ) and Sense of Coherence (SOC); at T2 (after the 20th weeks of pregnancy) they answered EDPS, SOC and MFA Scale (MFAS), and finally at T3 (6 months after birth), they repeated EDPS and SOC, and their interaction with their babies was videotaped and later evaluated using EA Scales. A statistical analysis has been done using descriptive statistics, correlation analysis, univariate logistic regression and multiple linear regression. Results The study has increased our knowledge on this particular population living in a multicultural, suburb community. It allow us to identify specific groups with a higher level of psychosocial risk, such as single or divorced women, young couples, mothers with a low level of education and those who are depressed or have a low SOC. The hypothesis that psychosocial risk is directly correlated with MFAS and that MFA is directly correlated with EA was not confirmed, neither the correlation between prenatal psychosocial risk and mother-baby EA. The study identified depression as a relevant risk factor in pregnancy and its higher prevalence in single or divorced women, immigrants and in those who have a higher global psychosocial risk. Depressed women have a poor MFA, and a lower structuring capacity and a higher hostility to their babies. In average, depression seems to reduce among pregnant women in the second part of their pregnancy. The children of immigrant mothers show a lower level of responsiveness to their mothers what could be transmitted through depression, as immigrant mothers have a higher risk of depression in the beginning of pregnancy and six months after birth. Young mothers have a low MFA and are more intrusive. Women who have a higher level of education are more sensitive and their babies showed to be more responsive. Women who are or have been submitted to abuse were found to have a higher level of MFA but their babies are less responsive to them. The study highlights the relevance of SOC as a potential protective factor while it is strongly and negatively related with a wide range of risk factors and mental health outcomes especially depression before, during and after pregnancy. Conclusions ALPHA proved to be a valid, feasible and reliable instrument to Primary Health Care (PHC) that can be used as a total sum score. We could not prove the association between psychosocial risk factors and MFA, neither between MFA and EA, or between psychosocial risk and EA. Depression and SOC seems to have a clear and opposite relevance on this process. Pregnancy can be considered as a maturational process and an opportunity to change, where adaptation processes occur, buffering risk, decreasing depression and increasing SOC. Further research is necessary to better understand interactions between variables and also to clarify a better operationalization of MFA. We recommend the use of ALPHA, SOC and EDPS in early pregnancy as a way of identifying more vulnerable women that will require additional interventions and support in order to decrease risk. At political level we recommend the reinforcement of Immigrant integration and the increment of education in women. We recommend more focus in health care and public health in mental health condition and psychosocial risk of specific groups at high risk. In PHC special attention should be paid to pregnant women who are single or divorced, very young, low educated and to immigrant mothers. This study provides the basis for an intervention programme for this population, that aims to reduce broad spectrum risk factors and to promote Mental Health in women who become pregnant. Health and mental health policies should facilitate the implementation of the suggested measures.

Correlação entre a cistatina C sérica e marcadores de aterosclerose subclínica em pacientes hipertensos

FUNDAMENTO: A cistatina C sérica (s-CC), um marcador endógeno da função renal, tem sido proposta também como um marcador de risco cardiovascular. No entanto, ainda não está estabelecido se se trata de um marcador direto de aterosclerose, independentemente da função renal. OBJETIVO: O objetivo deste estudo foi correlacionar a s-CC com dois marcadores substitutos de aterosclerose subclínica. MÉTODOS: Trata-se de um estudo transversal envolvendo 103 pacientes hipertensos ambulatoriais, de meia idade (57,49 ± 11,7 anos), sendo 60 do sexo feminino (58,25%) e a maioria com função renal preservada. A s-CC foi correlacionada com a espessura mediointimal carotídea (EMIc) e a dilatação mediada por fluxo de artéria braquial (DMF), ambas avaliadas por ultrassonografia, bem como com o clearance de creatinina medido e fatores de risco cardiovascular estabelecidos. RESULTADOS: A s-CC não se correlacionou significativamente nem com a EMIc (r = -0,024, p = 0,84) nem com a DMF (r = -0,050 e p = 0,687), e não foi observada também associação significativa com fatores de risco convencionais nem marcadores inflamatórios. Na análise univariada, a s-CC se correlacionou com o clearance de creatinina medido (r = - 0,498, p < 0,001), idade (r = 0,408, p < 0,001), microalbuminúria (r = 0,291, p = 0,014), ácido úrico (r = 0,391, p < 0,001), relação E/e' (r = 0,242, p = 0,049) e escore de Framingham (r = 0,359, p = 0,001). No entanto, após análise de regressão múltipla, apenas a associação com o clearance de creatinina medido permaneceu significativa (r = -0,491, p <0,001). CONCLUSÃO: Em pacientes hipertensos ambulatoriais de meia idade, a s-CC se correlacionou com o clearance de creatinina medido,como esperado, mas não foi observada associação com marcadores de aterosclerose nem com fatores de risco cardiovascular estabelecidos.

Produtividade da cana-de-açúcar em relação a clima e solos da região noroeste do estado de São Paulo

O trabalho baseou-se em seis experimentos que continham seis variedades comuns, instalados na região noroeste do estado de São Paulo. Em cada local, efetuou-se levantamento pedológico, seguido de amostragem para caracterização química, física e físico-hídrica em amostras deformadas e indeformadas, coletadas até à profundidade de aproximadamente 150 cm, por horizontes. Foram ainda coletados dados de clima e os rendimentos agrícolas referentes à cana-planta da safra 96/97. Para determinar os fatores do solo e do clima que melhor se relacionaram com a produtividade agrícola, foram feitas análises conjuntas, correlações e regressões lineares múltiplas para os diferentes conjuntos de dados de locais e variedades. As características químicas, físicas e hídricas da camada subsuperficial (B) apresentaram significativas correlações com o rendimento da cultura; o procedimento "stepwise" selecionou o Ca da camada B como a variável que mais se correlacionou com o rendimento da cana. O regime hídrico dos solos mostrou-se um fator importante para avaliação do potencial agrícola dos solos. Os Podzólicos eutróficos apresentaram maior capacidade potencial de produção na região considerada.

Avaliação de fatores edafoclimáticos e do manejo na produção de cafeeiros (Coffea arabica L.) no Oeste Paulista

Atributos do clima, solo e manejo foram caracterizados em 10 lavouras comerciais de café, estabelecidas na região oeste do estado de São Paulo, nos municípios de Sagres, Pompéia, Iacri, Bastos, Herculândia, Marília, Vera Cruz, Garça e Gália. Os objetivos foram identificar os principais fatores limitantes à cultura e estabelecer um modelo multivariado simples para explicar as produções observadas entre 1984 e 1989. Foram analisados, por correlação simples, mais de 2.000 atributos em relação às produções finais. Os atributos que apresentaram significância a 5% de probalidade pelo teste t de Student foram combinados em sucessivas análises de regressão múltipla. Três equações, com quatorze, cinco e seis variáveis preditoras, foram selecionadas, com graus de explicação da produção observada da ordem de 77, 73 e 75%, respectivamente. Quatro variáveis foram comuns aos três modelos: produção do ano anterior, idade da lavoura, temperatura mínima absoluta média na época do abotoamento e florescimento e soma dos teores de silte e argila nos horizontes subsuperficiais. Outras variáveis selecionadas foram: precipitação na época do abotoamento e florescimento; precipitação na época da maturação e colheita; densidade do solo e água disponível junto à cova (camada de 20-40 cm); macroporosidade na rodagem de máquinas (camada de 0-20 cm); adubação nitrogenada; pH em água e fósforo, ambos na área adubada (camada de 20-40 cm); teor de cálcio trocável e relação Mg:K trocáveis, ambos na área adubada (camada de 0-20 cm). A análise dos dados de produção no período de seis anos mostrou que as lavouras encontravam-se em processo de depauperamento, com declínio da produtividade da ordem de 400 kg ha-1 ano-1. As áreas necessitavam de melhoria da calagem, fosfatagem em profundidade, restabelecimento dos níveis de magnésio no solo e adubação mineral equilibrada. Também ficou evidenciada a necessidade de ser evitada a instalação da lavoura em áreas de acúmulo de ar frio no inverno ou solos com acentuado gradiente textural.