Joint Analysis of Multiple Algorithms and Performance Measures

There has been an increasing interest in the development of new methods using Pareto optimality to deal with multi-objective criteria (for example, accuracy and time complexity). Once one has developed an approach to a problem of interest, the problem is then how to compare it with the state of art. In machine learning, algorithms are typically evaluated by comparing their performance on different data sets by means of statistical tests. Standard tests used for this purpose are able to consider jointly neither performance measures nor multiple competitors at once. The aim of this paper is to resolve these issues by developing statistical procedures that are able to account for multiple competing measures at the same time and to compare multiple algorithms altogether. In particular, we develop two tests: a frequentist procedure based on the generalized likelihood-ratio test and a Bayesian procedure based on a multinomial-Dirichlet conjugate model. We further extend them by discovering conditional independences among measures to reduce the number of parameters of such models, as usually the number of studied cases is very reduced in such comparisons. Data from a comparison among general purpose classifiers is used to show a practical application of our tests.

Finite Population Inference for Causal Parameters

Thesis (Ph.D.)--University of Washington, 2016-08

Testing the random walk hypothesis with variance ratio statistics

Mestrado em Finanças

Power Study on Testing Epidemic Alternatives

Detecting change points in epidemic models has been studied by many scholars. Yao (1993) summarized five existing test statistics in the literature. Out of those test statistics, it was observed that the likelihood ratio statistic showed its standout power. However, all of the existing test statistics are based on an assumption that population variance is known, which is an unrealistic assumption in practice. To avoid assuming known population variance, a new test statistic for detecting epidemic models is studied in this thesis. The new test statistic is a parameter-free test statistic which is more powerful compared to the existing test statistics. Different sample sizes and lengths of epidemic durations are used for the power comparison purpose. Monte Carlo simulation is used to find the critical values of the new test statistic and to perform the power comparison. Based on the Monte Carlo simulation result, it can be concluded that the sample size and the length of the duration have some effect on the power of the tests. It can also be observed that the new test statistic studied in this thesis has higher power than the existing test statistics do in all of cases.

Assessing the fit of unidimensional IRT models for binary data under model misspecification

Model misspecification affects the classical test statistics used to assess the fit of the Item Response Theory (IRT) models. Robust tests have been derived under model misspecification, as the Generalized Lagrange Multiplier and Hausman tests, but their use has not been largely explored in the IRT framework. In the first part of the thesis, we introduce the Generalized Lagrange Multiplier test to detect differential item response functioning in IRT models for binary data under model misspecification. By means of a simulation study and a real data analysis, we compare its performance with the classical Lagrange Multiplier test, computed using the Hessian and the cross-product matrix, and the Generalized Jackknife Score test. The power of these tests is computed empirically and asymptotically. The misspecifications considered are local dependence among items and non-normal distribution of the latent variable. The results highlight that, under mild model misspecification, all tests have good performance while, under strong model misspecification, the performance of the tests deteriorates. None of the tests considered show an overall superior performance than the others. In the second part of the thesis, we extend the Generalized Hausman test to detect non-normality of the latent variable distribution. To build the test, we consider a seminonparametric-IRT model, that assumes a more flexible latent variable distribution. By means of a simulation study and two real applications, we compare the performance of the Generalized Hausman test with the M2 limited information goodness-of-fit test and the Likelihood-Ratio test. Additionally, the information criteria are computed. The Generalized Hausman test has a better performance than the Likelihood-Ratio test in terms of Type I error rates and the M2 test in terms of power. The performance of the Generalized Hausman test and the information criteria deteriorates when the sample size is small and with a few items.

[impact Variables On The Decline In Infant Mortality In The State Of São Paulo, Brazil: 1998-2008].

This is an ecological, analytical and retrospective study comprising the 645 municipalities in the State of São Paulo, the scope of which was to determine the relationship between socioeconomic, demographic variables and the model of care in relation to infant mortality rates in the period from 1998 to 2008. The ratio of average annual change for each indicator per stratum coverage was calculated. Infant mortality was analyzed according to the model for repeated measures over time, adjusted for the following correction variables: the city's population, proportion of Family Health Programs (PSFs) deployed, proportion of Growth Acceleration Programs (PACs) deployed, per capita GDP and SPSRI (São Paulo social responsibility index). The analysis was performed by generalized linear models, considering the gamma distribution. Multiple comparisons were performed with the likelihood ratio with chi-square approximate distribution, considering a significance level of 5%. There was a decrease in infant mortality over the years (p < 0.05), with no significant difference from 2004 to 2008 (p > 0.05). The proportion of PSFs deployed (p < 0.0001) and per capita GDP (p < 0.0001) were significant in the model. The decline of infant mortality in this period was influenced by the growth of per capita GDP and PSFs.

Clinicopathological And Immunohistochemical Evaluation Of Oral And Oropharyngeal Squamous Cell Carcinoma In Chilean Population.

In oral and oropharyngeal squamous cell carcinoma (OCSCC and OPSCC) exist an association between clinical and histopathological parameters with cell proliferation, basal lamina, connective tissue degradation and surrounding stroma markers. We evaluated these associations in Chilean patients. A convenience sample of 37 cases of OCSCC (n=16) and OPSCC (n=21) was analyzed clinically (TNM, clinical stage) and histologically (WHO grade of differentiation, pattern of tumor invasion). We assessed the expression of p53, Ki67, HOXA1, HOXB7, type IV collagen (ColIV) and carcinoma-associated fibroblast (α-SMA-positive cells). Additionally we conducted a univariate/bivariate analysis to assess the relationship of these variables with survival rates. Males were mostly affected (56.2% OCSCC, 76.2% OPSCC). Patients were mainly diagnosed at III/IV clinical stages (68.8% OCSCC, 90.5% OPSCC) with a predominantly infiltrative pattern invasion (62.9% OCSCC, 57.1% OPSCC). Significant association between regional lymph nodes (N) and clinical stage with OCSCC-HOXB7 expression (Chi-Square test P < 0.05) was observed. In OPSCC a statistically significant association exists between p53, Ki67 with gender (Chi-Square test P < 0.05). In OCSCC and OPSCC was statistically significant association between ki67 with HOXA1, HOXB7, and between these last two antigens (Pearson's Correlation test P < 0.05). Furthermore OPSCC-p53 showed significant correlation when it was compared with α-SMA (Kendall's Tau-c test P < 0.05). Only OCSCC-pattern invasion and OPSCC-primary tumor (T) pattern resulted associated with survival at the end of the follow up period (Chi-Square Likelihood Ratio, P < 0.05). Clinical, histological and immunohistochemical features are similar to seen in other countries. Cancer proliferation markers were associated strongly from each other. Our sample highlights prognostic value of T and pattern of invasion, but the conclusions may be limited and should be considered with caution (small sample). Many cases were diagnosed in the advanced stages of the disease, which suggests that the diagnosis of OCSCC and OPSCC is made late.

Association Between Unilateral Tonsillar Enlargement And Lymphoma In Children: A Systematic Review And Meta-analysis.

Lymphoma is the most common head and neck malignancy in children, and palatine tonsils asymmetry is the most frequent clinical manifestation of tonsillar lymphoma. However, several studies with children with tonsillar asymmetry found no case of lymphoma, showing that the relationship of tonsillar asymmetry with lymphoma is unclear. In this review, we aimed to identify the association between tonsillar asymmetry and tonsillar lymphoma in children by conducting systematic reviews of the literature on children with palatine tonsil lymphoma and tonsillar asymmetry. Articles comprising the paediatric age group (up to 18 years) with information concerning clinical manifestations of tonsillar lymphoma or the diagnosis of the tonsillar asymmetry were included. The main cause of asymmetry of palatine tonsils was lymphoid hyperplasia, followed by lymphoma and nonspecific benign changes. The asymmetry of tonsils was present in 73.2% of cases of lymphoma. There was an association between asymmetric palatine tonsils and lymphoma, with a likelihood ratio of 43.5 for children with asymmetry of palatine tonsils and 8938.4 for children with asymmetry of tonsils and other signs of suspicion for malignancy. We also provide recommendations on the management of suspicious cases of palatine tonsil lymphoma.

Inclusão da epistasia em modelo de avaliação genética de bovinos de corte compostos

Dados de bovinos compostos foram analisados para avaliar o efeito da epistasia nos modelos de avaliação genética. As características analisadas foram os pesos aos 205 (P205) e 390 dias (P390) e perímetro escrotal aos 390 dias (PE390). As análises foram realizadas pela metodologia de máxima verossimilhança considerando-se dois modelos: o modelo 1 incluiu como covariáveis os efeitos aditivos diretos e maternos, e os não aditivos das heterozigoses para os efeitos diretos e para o materno total, e o modelo 2 considerou também o efeito direto de epistasia. Para comparação dos modelos, foram utilizados o critério de informação de Akaike (AIC) e o critério de informação Bayesiano de Schwartz (BIC), e o teste de razão de verossimilhança. A inclusão da epistasia no modelo de avaliação genética pouco alterou as estimativas de componentes de (co)variâncias genéticas aditivas e, consequentemente, as herdabilidades. O teste de verossimilhança e o critério de Akaike sugeriram que o modelo 2, que inclui a epistasia, apresentou maior aderência aos dados para todas as características analisadas. O critério BIC indicou este modelo como o melhor apenas para P205. Para análise genética dessa população, o modelo que considerou o efeito de epistasia foi o mais adequado.

Variabilidade amostral das séries mensais de precipitação pluvial em duas regiões do Brasil: Pelotas-RS e Campinas-SP

O presente trabalho avaliou a variabilidade amostral dos parâmetros da distribuição gama, relativos a séries mensais de precipitação pluvial, nas regiões de Campinas-SP e Pelotas-RS, que têm dados para os períodos de 1890-2006 e 1890-2005, respectivamente. Assim, os espaços amostrais considerados foram de 58, 39 e 29 anos para Campinas e 58 e 29 anos para Pelotas. As análises foram feitas usando o teste da razão da máxima verossimilhança. Os resultados apontaram significativas alterações amostrais. Não houve indicações de tendências contínuas (redução ou aumento) no regime mensal de precipitação pluvial na região de Campinas-SP. Em contrapartida, esse teste indicou, para a localidade de Pelotas-RS, tendência de elevação no regime desse elemento meteorológico ocorrida entre as amostras relativas aos anos de 1948 a 1976 e 1977 a 2005.

Prevalência do HPV em mulheres rastreadas para o câncer cervical

OBJETIVO: Analisar a prevalência da infecção genital por papilomavírus humano (HPV) de alto risco por faixa etária e fatores associados. MÉTODOS: Estudo transversal com amostra de 2.300 mulheres (15-65 anos) que buscaram rastreamento para o câncer cervical entre fevereiro de 2002 e março de 2003 em São Paulo e Campinas, estado de São Paulo. Aplicou-se questionário epidemiológico e realizou-se coleta cervical para citologia oncológica e teste de captura híbrida II. As análises estatísticas empregadas foram teste de qui-quadrado de Pearson e análise multivariada pelo método forward likelihood ratio. RESULTADOS: A prevalência total da infecção genital por HPV de alto risco foi de 17,8%, distribuída nas faixas etárias: 27,1% (<25 anos), 21,3% (25-34 anos), 12,1% (35-44 anos), 12,0% (45-54 anos) e de 13,9% (55-65 anos). Participantes com maior número de parceiros sexuais durante a vida apresentaram maior freqüência da infecção. Relacionamento estável, idade de 35 a 44 anos e ex-fumantes foram associados à proteção da infecção. A infecção genital por HPV de alto risco ocorreu em 14,3% das citologias normais, em 77,8% das lesões escamosas de alto grau e nos dois (100%) casos de carcinoma. CONCLUSÕES: A prevalência da infecção genital por HPV de alto risco na amostra estudada foi alta. Houve predomínio de casos abaixo dos 25 anos e tendência a um novo aumento após os 55 anos, com maior freqüência naqueles com maior número de parceiros sexuais durante a vida

Comparison of different models to estimate genetic parameters for carcass traits in a commercial broiler line

The effect of genetic and non-genetic factors for carcass, breast meat and leg weights, and yields of a commercial broiler line were investigated using the restricted maximum likelihood method, considering four different animal models, including or excluding maternal genetic effect with covariance between direct and maternal genetic effects, and maternal permanent environmental effect. The likelihood ratio test was used to determine the most adequate model for each trait. For carcass, breast, and leg weight, and for carcass and breast yield, maternal genetic and permanent environmental effects as well as the covariance between direct and maternal genetic effects were significant. The estimates of direct and maternal heritability were 0.17 and 0.04 for carcass weight, 0.26 and 0.06 for breast weight, 0.22 and 0.02 for leg weight, 0.32 and 0.02 for carcass yield, and 0.52 and 0.04 for breast yield, respectively. For leg yield, maternal permanent environmental effect was important, in addition to direct genetic effects. For that trait, direct heritability and maternal permanent environmental variance as a proportion of the phenotypic variance were 0.43 and 0.02, respectively. The results indicate that ignoring maternal effects in the models, even though they were of small magnitude (0.02 to 0.06), tended to overestimate direct genetic variance and heritability for all traits.

Plasma Superoxide Dismutase-1 as a Surrogate Marker of Vivax Malaria Severity

Background: Severe outcomes have been described for both Plasmodium falciparum and P. vivax infections. The identification of sensitive and reliable markers of disease severity is fundamental to improving patient care. An intense pro-inflammatory response with oxidative stress and production of reactive oxygen species is present in malaria. Inflammatory cytokines such as tumor necrosis factor-alpha (TNF-alpha) and antioxidant agents such as superoxide dismutase-1 (SOD-1) are likely candidate biomarkers for disease severity. Here we tested whether plasma levels of SOD-1 could serve as a biomarker of severe vivax malaria. Methodology/Principal Findings: Plasma samples were obtained from residents of the Brazilian Amazon with a high risk for P. vivax transmission. Malaria diagnosis was made by both microscopy and nested PCR. A total of 219 individuals were enrolled: non-infected volunteers (n = 90) and individuals with vivax malaria: asymptomatic (n = 60), mild (n = 50) and severe infection (n = 19). SOD-1 was directly associated with parasitaemia, plasma creatinine and alanine amino-transaminase levels, while TNF-alpha correlated only with the later enzyme. The predictive power of SOD-1 and TNF-alpha levels was compared. SOD-1 protein levels were more effective at predicting vivax malaria severity than TNF-alpha. For discrimination of mild infection, elevated SOD-1 levels showed greater sensitivity than TNF-alpha (76% vs. 30% respectively; p < 0.0001), with higher specificity (100% vs. 97%; p < 0.0001). In predicting severe vivax malaria, SOD-1 levels exhibited higher sensitivity than TNF-alpha (80% vs. 56%, respectively; p < 0.0001; likelihood ratio: 7.45 vs. 3.14; p, 0.0001). Neither SOD-1 nor TNF-alpha could discriminate P. vivax infections from those caused by P. falciparum. Conclusion: SOD-1 is a powerful predictor of disease severity in individuals with different clinical presentations of vivax malaria.

Anti-Anopheles darlingi saliva antibodies as marker of Plasmodium vivax infection and clinical immunity in the Brazilian Amazon

Background: Despite governmental and private efforts on providing malaria control, this disease continues to be a major health threat. Thus, innovative strategies are needed to reduce disease burden. The malaria vectors, through the injection of saliva into the host skin, play important role on disease transmission and may influence malaria morbidity. This study describes the humoral immune response against Anopheles (An.) darlingi saliva in volunteers from the Brazilian Amazon and addresses the association between levels of specific antibodies and clinical presentation of Plasmodium (P.) vivax infection. Methods: Adult volunteers from communities in the Rondonia State, Brazil, were screened in order to assess the presence of P. vivax infection by light microscopy and nested PCR. Non-infected volunteers and individuals with symptomatic or symptomless infection were randomly selected and plasma collected. An. darlingi salivary gland sonicates (SGS) were prepared and used to measure anti-saliva antibody levels. Plasma interleukin (IL)-10 and interferon (IFN)-gamma levels were also estimated and correlated to anti-SGS levels. Results: Individuals infected with P. vivax presented higher levels of anti-SGS than non-infected individuals and antibody levels could discriminate infection. Furthermore, anti-saliva antibody measurement was also useful to distinguish asymptomatic infection from non-infection, with a high likelihood ratio. Interestingly, individuals with asymptomatic parasitaemia presented higher titers of anti-SGS and lower IFN-gamma/IL-10 ratio than symptomatic ones. In P. vivax-infected asymptomatic individuals, the IFN-gamma/IL-10 ratio was inversely correlated to anti-SGS titers, although not for while in symptomatic volunteers. Conclusion: The estimation of anti-An. darlingi antibody levels can indicate the probable P. vivax infection status and also could serve as a marker of disease severity in this region of Brazilian Amazon.

Identification of microsatellites markers associated with yield components and quality parameters in sugarcane

Marker assisted selection depends on the identification of tightly linked association between marker and the trait of interest. In the present work, functional (EST-SSRs) and genomic (gSSRs) microsatellite markers were used to detect putative QTLs for sugarcane yield components (stalk number, diameter and height) and as well as for quality parameters (Brix, Pol and fibre) in plant cane. The mapping population (200 individuals) was derived from a bi-parental cross (IACSP95-3018 x IACSP93-3046) from the IAC Sugarcane Breeding Program. As the map is under construction, single marker trait association analysis based on the likelihood ratio test was undertaken to detect the QTLs. Of the 215 single dose markers evaluated (1:1 and 3:1), 90 (42%) were associated with putative QTLs involving 43 microsatellite primers (18 gSSRs and 25 EST-SSRs). For the yield components, 41 marker/trait associations were found: 20 for height, 6 for diameter and 15 for stalk number. An EST-SSRs marker with homology to non-phototropic hypocotyls 4 (NPH4) protein was associated with a putative QTL with positive effect for diameter as also with a negative effect for stalk number. In relation to the quality parameters, 18 marker trait associations were found for Brix, 19 for Pol, and 12 for fibre. For fibre, 58% of the QTLs detected showed a negative effect on this trait. Some makers associated with QTLs with a negative effect for fibre showed a positive effect for Pol, reflecting the negative correlation generally observed between these traits.