ABSTRACT. The identification and interpretation of discrepant observations in sensory experiments are difficult to implement since the external effects are associated with the individual consumer. This fact becomes more relevant in experiments that involve blends, which scrutinize coffees with different qualities, varieties, origins, and forms of processing and preparation. This work proposes a statistical procedure that facilitates the identification of outliers while also evaluating the discriminatory powers of a sensory panel concerning the differentiation of pure blends and coffees. For this purpose, four experiments were performed that tested coffees with different qualities and varieties. The results suggest that the statistical procedure proposed in this work was effective for discriminating the blends relative to the pure coffees and that the effects of the concentrations and types of processing did not interfere with the statistical evaluations.

Vários tipos de resíduos têm sido propostos para modelos de sobrevivência, sendo os mais adequados resultado dos resíduos generalizados de Cox e Snell (1968). O objetivo com este trabalho é avaliar a adequacidade de modelos por meio de gráficos de diagnósticos gerados a partir dos resíduos generalizados de Cox-Snell. Para ilustrar a teoria, foram feitas três aplicações. A primeira aplicação visou a ilustrar a lógica existente entre a plotagem dos resíduos ordenados de três distribuições, normal (0,1), logística (0,1) e valor extremo (0,1) versus as estatísticas de ordem esperadas desses resíduos de acordo com as distribuições assumidas. Para a segunda aplicação, foram utilizados dados de tempo de vida de isolantes, obtidos em Nelson (1990). A partir da verificação por meio dos gráficos de diagnósticos utilizando-se os resíduos generalizados de Cox-Snell, encontrou-se que o modelo apropriado para o tempo de vida dos isolantes era o log-normal. Para a terceira aplicação, foram analisados dados censurados referentes ao tempo de vida de pacientes, obtidos em Collett (1994). Avaliou-se a adequacidade de vários modelos por meio dos resíduos de Cox-Snell adaptados para dados de sobrevivência. Pelos resultados constatou-se que o modelo Weibull foi o mais adequado.

Several kinds of residuals have been proposed for survival models, the most suitable for this purpose are Cox and Snell (1968) generalized residuals. The objective of this work was to evaluate the adequacy of models by graphical diagnostics using Cox-Snell generalized residuals. To illustrate the theory three applications were considered. The first application sought to illustrate the heuristics by plotting ordered residuals from three distributions: normal (0,1), logistics (0,1) and extreme value (0,1), versus the expected order statistics of these residuals in consonance with the assumed distributions. The second application consisted of lifetime data of electric insulating, obtained by Nelson (1990). Starting from graphical diagnostics using Cox-Snell generalized residuals, it was found that the model appropriate for lifetime of electric insulating was log-normal. The third application referred to censored data of lifetime of patients, obtained by Collett (1994). The adequacy of several models was evaluated by Cox-Snell generalized residuals adapted for survival data. The results show that Weibull model was the most appropriate.