Evaluation of Surival models using censored R Package

[Che_Hidayat_Che_nawi]

Im practicing survival machine learning modeling using censored R package. However, I observe that there is no features available for model evaluation. I appreciate if anyone can assist me with model evaluation.

Thank you

[Max]

We are almost done testing and documenting the functions and methods to evaluate these models.

Evaluation is fairly complicated due to censoring. We've decided to use inverse probability censoring weights (ipcw) to account for that. A good referene for that is

Graf, E., Schmoor, C., Sauerbrei, W. and Schumacher, M. (1999), Assessment and comparison of prognostic classification schemes for surviva data. Statist. Med., 18: 2529-2545.

Also, there are static metrics, dynamic metrics, and integrated metrics.

Based on all of this, we have to write a lot of documentation since it is complex. We can just say "use funciton foo()". We're working on that and we will post in this section of Community (and blog posts etc) once we have documentation that explains it all.

[Che_Hidayat_Che_nawi]

Thank you for your feedback.

[arjunsuhass]

The censored R Package lacks model evolution features. The admin is testing methods for parking master multiplayer 2 download, such as inverse probability censoring weights, and working on documentation to explain the complex topic.