Heliton Ribeiro Tavares

Item response theory for longitudinal data: Item and population ability parameters estimation

In this paper we deal with the problem of estimating item and population ability parameters in longitudinal studies where a group of individuals is submitted to different tests along the time, with some common items. Several covariance structures are explored to model the possible dependency between the abilities of the same individual, measured at different instants. Maximum likelihood equations and some simulation results are presented.

Competitividade inter-regional de sistemas de produção de bovinocultura de corte

This paper presents a methodology for evaluating production systems by measuring the intern competitiveness of in beef cattle. During the first quarter of 2010, questionnaires were administered to 65 farmers, 36 interviews in the Southern Region (State of Rio Grande do Sul) and 29 in the Northern Region (States of Para and Rondonia). Defined the key drivers that affect the international competitiveness - technology, management, market relations and institutional environment, an specific weight was assigned to each in order to get the index of competitiveness. The results were statistically analyzed by the item response theory and the correspondence analysis (ANACOR) with the SPSS® software. The South was more competitive than the North. Regardless of region, the critical factors of competitiveness were: crop-livestock integration, strategic planning, calculation of financial indicators, beef pricing, access to technological innovations and organization of farmers.

MODELO DE EFEITOS FIXOS COM MEDIDA REPETIDA APLICADO EM EXPERIMENTOS DE MELHORAMENTO GENÉTICO DO CUPUAÇUZEIRO

This study was conducted to evaluate vegetative performance of cupuassu progenies, in the immaturity phase, through analysis of repeated measures. The experimental area was installed in the municipality of Tome Acu, Northeast of the state of Para. We employed 25 full-sib progeny of cupuassu, who had vegetative growth (height and diameter of the plant), monitored for three years. Because of the nature of longitudinal observations it was assessed primarily through Mauchly sphericity test, which type of statistical analysis should be applied. For the variable diameter, sphericity conditionwas not violated, therefore, proceeded to review the design scheme of split plot. For the height variable, the test was significant, indicating necessarily evaluate the best structure to explain the correlation of errors, being chosen the covariance structure Heterogeneous Composed Symmetry. Differences were found between the progenies only in the third year of evaluation, and the variable diameter of the plant allowed better discriminate the progenies.