Competing Risk Analysis of Event History Data Using Recurrent Neural Networks

DOI

Bibliographic Information

Other Title
  • リカレントニューラルネットワークを活用したイベントヒストリーデータの競合リスク解析

Search this article

Description

Bone marrow transplantation data with multistate process have competing risks in which an individual can experience two types of death, that is death due to original disease and competing risks. Multi-value type recurrent neural networks are proposed based on softmax transformation. We can select the optimum hidden unit based on bootstraping. Outliers are identified by using influential analysis. The deviance on fitting of the recurrent neural network can be bootstrapped. We can estimate the proportions of discrete hazards and survival probability during the course of the disease with better accuracy than multi-value type feed-forward neural networks.

Journal

Keywords

Details 詳細情報について

Report a problem

Back to top