## Math & Physical Sciences Posters

#### Presentation Title

Bayesian Analysis of Rat Survival Times Under a Proportional Mean Residual Life Model

Valerie Poynor

#### Start Date

17-11-2018 12:30 PM

#### End Date

17-11-2018 2:30 PM

CREVELING 120

POSTER 2

Poster

#### Subject Area

physical_mathematical_sciences

#### Abstract

Survival data describes time to a particular event such as time to death of a patient, time to failure of a machine, or time to injury-recovery of an athlete. In our research, we consider a data set describing the survival times of rats under two experimental groups. In the study, the rats were divided randomly into the two groups: one in which they were allowed to eat freely in quantity and time (Adlibitum) and the other in which food was restricted in quantity and time (Restricted). In our analysis, we compare the mean residual life function of the two groups. In survival analysis, the mean residual life function describes the expected remaining life given survival up to a particular time, t. This function is desirable in both theory and application. Namely, the function characterizes the survival distribution while offering a practical interpretation. We apply a Bayesian proportional mean residual life function to the data. Our model assumes a baseline mean residual life function that is associated with the gamma distribution. We provide point and interval estimates for the survival and mean residual life functions and compare our results with an Exponentiated Weibull model.

Keywords: Proportional mean residual life, Bayesian modeling, survival analysis, Monte Carlo Markov chain

#### Share

COinS

Nov 17th, 12:30 PM Nov 17th, 2:30 PM

Bayesian Analysis of Rat Survival Times Under a Proportional Mean Residual Life Model

CREVELING 120

Survival data describes time to a particular event such as time to death of a patient, time to failure of a machine, or time to injury-recovery of an athlete. In our research, we consider a data set describing the survival times of rats under two experimental groups. In the study, the rats were divided randomly into the two groups: one in which they were allowed to eat freely in quantity and time (Adlibitum) and the other in which food was restricted in quantity and time (Restricted). In our analysis, we compare the mean residual life function of the two groups. In survival analysis, the mean residual life function describes the expected remaining life given survival up to a particular time, t. This function is desirable in both theory and application. Namely, the function characterizes the survival distribution while offering a practical interpretation. We apply a Bayesian proportional mean residual life function to the data. Our model assumes a baseline mean residual life function that is associated with the gamma distribution. We provide point and interval estimates for the survival and mean residual life functions and compare our results with an Exponentiated Weibull model.

Keywords: Proportional mean residual life, Bayesian modeling, survival analysis, Monte Carlo Markov chain