Modelo de regressão odds-riscos proporcionais para dados de sobrevivência discretos

Content Provider	Semantic Scholar
Author	Vieira, Maria Gabriella Figueiredo
Copyright Year	2018
Abstract	There is an extensive literature that develops the Survival Analysis techniques to continuous failure times, mainly as regards the adjustment of regression models. The use of continuous models to analyze a discrete variable is not acceptable in certain situations. Thus, since the theoretical reference is restricted, the development of studies in this area becomes very relevant. Therefore, the purpose of this work is to structure a regression model for survival data with discrete failure time. For this, based on the Cox proportional hazards regression model, an analog model is proposed that adapts the theory so that the covariables act multiplicatively in the risk odds. As result, a technique of regression model adjustment for discrete data is obtained, proposing a new way to adjust the model, to verify the assumptions and to analyze the adjustment quality. Using the database of a study done with patients who feel low back pain, the proportional odds-risks regression model is adjusted and the whole proposed analysis is carried out. It is concluded that the proportional odds-risks regression model is a good and complete technique to analyze the variables that are originally discrete.
File Format	PDF HTM / HTML
Alternate Webpage(s)	http://repositorio.unb.br/bitstream/10482/34741/1/2018_MariaGabriellaFigueiredoVieira.pdf
Language	English
Access Restriction	Open
Content Type	Text
Resource Type	Article