Abstract
In 28 patients with metastatic colonic cancer the treatment modality was a significant predictor of event with a p-value of 0,010 in the Cox regression. With the Accelerated failure time models the best predictive p-value was p = 0,001 which was ten times better than that of the Cox model. For all of the AFT (Accelerated failure time) models the AIC (Akaike Information Criterion) goodness of fit measure was over twice better with AIC values around 25,000 as compared to that of the Cox regression of close to 70,000 (the smaller the value, the better the fit). AFT p-values were correspondingly generally much smaller than those of the Cox regression.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
