Development and Modeling of Decision Tree for Survival Data with Multiple Events Using Deviance and Cox-Snell Residuals within Node Homogeneity Technique

It is very common in medical studies for a patient to experience more than one event rather of interest. This led exposing an individual multiple risks and practitioners need account these concerning some prognostic factors. There are many methods dealing with events survival data classically, however, break down when considering the top-down effect factors concurrently correlated (competing risks). study aimed develop decision tree using within-node homogeneity procedure analysis classify competing risks. Since CART methodology involves recursive portioning covariates into different subgroups, this considers use Deviance Modified Cox-Snell residuals as measure impurity Classification Regression Tree (CART) during process partitioning. The flexibility predictive accuracy our learning algorithm would then be compared other existing through simulation freely available online real-life data. results revealed that: response within node classification performs better irrespective performance indices. Results from empirical two that proposed model residual (Deviance=16.6498) both Martingale (deviance=160.3592) (Deviance=556.8822). Conclusively, (Mean Square Error (MSE)=0.01783563) improved any (MSE=0.1853148, 0.8043366). implies have capability accounting effects based on biomarkers.