2018-04-242017-02-03MIRANDA, Isabela Pagani Heringer de. Modelos de medidas repetidas aplicados a um experimento de pacientes com câncer tratados com exergames. 2017. 78 f. Dissertação (Mestrado em Estatística Aplicada e Biometria) - Universidade Federal de Alfenas, Alfenas, MG, 2017.https://repositorio.unifal-mg.edu.br/handle/123456789/1134The objective was to use repeated measures techniques to define the appropriate linear model that is able to compare three groups of individuals regarding the methods applied in the reduction of muscular fatigue. The cancer group consisted of individuals diagnosed with cancer who were already treated with chemotherapy and radiotherapy, the control group was composed of healthy individuals and the chemo / radio group was composed of individuals diagnosed with cancer undergoing chemotherapy and radiotherapy. Study participants were treated with exergames. Each individual obtained a series of muscle strength data in six muscles (right/left medial gastrocnemius, right/left lateral gastrocnemius e right/left anterior tibialis) and in three sessions (pre treatment, half treatment and end treatment). The groups, the sessions, the measurements, as well as the double interactions and the triple interaction between the factors were analyzed according to the splits plots designs in time using linear mixed models, which imposes strong restriction regarding the covariance matrix of the experimental error. According to the results of the Mauchly test, the sphericity hypothesis of the covariance matrix was rejected for all six muscles, so we chose the structure of the covariance matrix of the experimental error that best fits the data. The structure ARMA(1.1) was the most adequate structure according to the AIC and BIC criteria. Considering the structure ARMA(1,1) in the MIXED procedure in the software SAS, the analysis of variance for the muscle strength data in each muscle under study was performed. Only for the left lateral gastrocnemius muscle none of the factors was significant. For the other muscles, there were significant differences and tendencies in the data series, especially for the left tibial muscle, in which the interactions between group and session and between group and measurement were significant, showing that exergamen treatment increased muscle strength in debilitated patients and, With 20 treatment sessions, the groups matched for muscle strength.application/pdfAcesso Abertohttp://creativecommons.org/licenses/by-nc-nd/4.0/Modelos LinearesAnálise de VariânciaCIENCIAS AGRARIASDissertaçãoNogueira, Denismar Alves