On the use of Harrell’s C for clinical risk prediction via random survival forests
作者：
Highlights：
• Harrell’s C is proposed as a split criterion in random survival forests.
• Split points of continuous predictor variables differ substantially between Harrell’s C and log-rank splitting.
• The log-rank statistic has a stronger end-cut preference than Harrell’s C.
• Harrell’s C outperforms log-rank splitting in smaller scale studies.
• Harrell’s C outperforms log-rank splitting if the censoring rate is high.
摘要
•Harrell’s C is proposed as a split criterion in random survival forests.•Split points of continuous predictor variables differ substantially between Harrell’s C and log-rank splitting.•The log-rank statistic has a stronger end-cut preference than Harrell’s C.•Harrell’s C outperforms log-rank splitting in smaller scale studies.•Harrell’s C outperforms log-rank splitting if the censoring rate is high.
论文关键词：Concordance index,Event history analysis,Log-rank statistic,Random survival forests,Risk prediction,Split rules
论文评审过程：Received 13 March 2016, Revised 11 July 2016, Accepted 12 July 2016, Available online 16 July 2016, Version of Record 22 July 2016.
论文官网地址：https://doi.org/10.1016/j.eswa.2016.07.018
