Navegando por Data de Publicação, começando com "2019-04-01"

Filtrar resultados por ano ou mês

Agora exibindo 1 - 3 de 3

Um procedimento para seleção de variáveis em modelos lineares generalizados duplos
(Universidade Federal de São Carlos, 2019-04-01) Cavalaro, Lucas Leite; Pereira, Gustavo Henrique de Araujo; https://lattes.cnpq.br/4536501674241631; https://lattes.cnpq.br/2371167969495052
The double generalized linear models (DGLM), unlike the generalized linear model (GLM), allow the fit of the dispersion parameter of the response variable as a function of predictor variables, improving the way of modeling phenomena. Thus, they are a possible solution when the assumption that the constant dispersion parameter is unreasonable and the response variable has distribution belonging to the exponential family. Considering our interest in variable selection in this class of models, we studied the two-step variable selection scheme proposed by Bayer and Cribari-Neto (2015) and, based on this method, we developed a scheme to select variables in up to “k” steps. To check the performance of our procedure, we performed Monte Carlo simulation studies in DGLM. The results indicate that our procedure for variable selection presents, in general, similar or superior performance than the other studied methods without requiring a large computational cost. We also evaluated the scheme to select variables in up to “k” steps in a set of real data and compared it with different regression methods. The results showed that our procedure can also be a good alternative when the interest is in making predictions.
Habilidades sociais educativas de docentes universitários na avaliação de alunos: validação de um instrumento
(Universidade Federal de São Carlos, 2019-04-01) Santos, Joene Vieira dos; Del Prette, Almir; https://lattes.cnpq.br/2191226494887247; https://lattes.cnpq.br/2758607421177423
The present study had as objectives to examine: (a) evidence of validity of the Inventory of Educational Social Skills of the University Professor – Student Version (IESS-UPStudent) based on the internal structure and relationship with the satisfaction criterion variable; (b) the relationship between teacher and student evaluation of the professor’s HSE repertoire; and (c) the differences in the HSE level attributed to the teacher according to the characteristics of the sample. One thousand four hundred and six students and 16 university professors participated in this study. The results showed that IESS-UP-Student is composed of five dimensions: approve and value student behavior (AVB, 13 items), expose, explain and evaluate interactively (EEE, 11 items), cultivate affection, support and good humor (ASG, 10 items), reject undesirable student behaviors (RUB, 9 items) and guide activities (GAc, 6 items), which form a higher dimension called the university professor’s ESS. The invariance analysis of the model showed that this structure is equivalent in function of characteristics of the professor, student and Institution of Higher Education (IHE). The data also indicated that teacher’s HSE interferes positively and significantly in the level of student satisfaction with the teaching performance (variable criterion) and there are discrepancies between teachers’ and students’ perceptions of professor performance. In addition, differences in the HSE level attributed to teachers were identified according to the characteristics of the teacher (gender, role and contact in other disciplines), student (gender, area of knowledge and moment of the course) and IHE (nature, region and philosophy institutional). These results seem to indicate that the IESS-UP-Student can be a useful tool to: (a) examine the learning conditions offered to the student, based on the professor’s HSE; (b) provide feedback to the teacher about the relationship established with their students; and (c) to evaluate the effectiveness of intervention programs aimed at developing the HSE of college teachers.
Curvas de rotação galácticas em teorias métricas da gravitação
(Universidade Federal de São Carlos, 2019-04-01) Gauy, Henrique Matheus; Ramos-Caro, Javier Fernando; https://lattes.cnpq.br/2067343302764973; https://lattes.cnpq.br/1412367289023716
From a Parametrized Post-Newtonian (PNP) perspective we attempt to partially answer the question, by studying some models, of whether or not the new degrees of freedom, represented by the PNP potentials, can lead to significant modifications in the dynamics of galaxies in the direction of rendering dark matter obsolete. In the PNP, several parameters classify each theory and their possible values are tightly constrained mainly due to solar system experiments [ 1 , 2 , 3 ]. Such restrictions renders the modifications of gravitational effects, with respect to General Relativity, to be insignificant, making attempts to find alternative metrical theories rather fruitless. In recent years a new kind of metric theories have arisen to popularity and are characterized by "Screening" mechanisms [ 4 ]. These lead to the possibility of modifications in larger scales than the solar system while retaining the success of GR in the latter, allowing for violations of the constraints of the Post-Newtonian parameters. With this new paradigm we consider two kinds of solutions: (i) When no matter field is present, or in other words, vacuum; (ii) or a Politropic distribution of matter. For (i) we find that the corrections don’t lead to the needed modifications as required to be an alternative to dark matter. In (ii) for most choices of the parameters (when changing γ , β , ξ , α 3 , ζ j , with j = 1 , 2, from their values in GR) the need for dark matter is unavoidable, in order to find a flat rotation curve. It’s only for theories in which ζ 3 > 0 that some resemblance of flat rotation curves is found. The latter suggests, at least for the models considered, that these are the only theories capable of replacing dark matter as a possible explanation for the dynamics of galaxies.