Softwares

When data cannot cross borders

Comparing models fitted in different countries This example is based on a presentation at the Association of Nordic Cancer Registries Conference 2024. The presentation can be found here International collaborative research using cancer registry data enables exploration of differences in cancer incidence, mortality, and survival.

Introduction to mlad - a Weibull survival model

mlad

mlad - maximizing likelihood functions using automatic differentiation

mlad maximizes a log-likelihood function where the log-likelihood function is programmed in Python. This enables the gradients and Hessian matrix to be obtained using automatic differentiation and makes better use of multiple CPUs.

Using mlad for interval censored data

mlad

Using mlad for splines on the log hazard scale.

mlad

Numerical Integration for log hazard models

stpm3

Change of scale option

stpm3

Contrasts

stpm3

Extended Functions

stpm3

Factor Variables

stpm3