Novel Maximum Entrophy Method for the Average Survival Time Differences between Two Groups
We suggest a Maximum Likelihood estimator for the Average Survival Time Difference between two groups, based on the extra time in which should be added to one group to produce the maximum entropy of the result. The estimator is calculated only from time to event data, does not assume hazard proportionality and provides directly the magnitude of the clinical differences between the groups. Our Monte Carlo simulations show that, even at low sample numbers the estimator is a reliable predictor for the clinical differences between the groups, and therefore can be used to estimate from preliminary data whether or not the large sample number experiment is worth pursuing.
Guzman-Gonzalez, Joscelyne, "Novel Maximum Entrophy Method for the Average Survival Time Differences between Two Groups" (2019). ETD Collection for University of Texas, El Paso. AAI22618400.