方法对比
并排查看您选择的方法；存在差异的行会高亮显示。
| 贝叶斯 Cox 回归× | 贝叶斯生存回归× | |
|---|---|---|
| 领域 | 统计学 | 统计学 |
| 方法族 | Regression model | Regression model |
| 起源年份≠ | 1972 (Cox PH); 2001 (Bayesian treatment) | 1990s–2001 |
| 提出者≠ | Cox (1972) for the base model; Bayesian formulation by Sinha, Chen & Ghosh (1990s); comprehensive treatment by Ibrahim, Chen & Sinha (2001) | Ibrahim, Chen & Sinha (seminal textbook treatment, 2001); broader Bayesian framework: Gelman et al. |
| 类型≠ | Survival regression | Bayesian parametric/semiparametric regression |
| 开创性文献 | Ibrahim, J. G., Chen, M.-H., & Sinha, D. (2001). Bayesian Survival Analysis. Springer. ISBN: 978-0387952772 | Ibrahim, J. G., Chen, M.-H., & Sinha, D. (2001). Bayesian Survival Analysis. Springer. ISBN: 978-0387952772 |
| 别名 | Bayesian Cox PH model, Bayesian proportional hazards model, Bayesian survival regression, BCox | Bayesian time-to-event regression, Bayesian parametric survival model, Bayesian survival analysis, Bayesian accelerated failure time model |
| 相关≠ | 6 | 5 |
| 摘要≠ | Bayesian Cox regression combines the Cox proportional hazards model for time-to-event data with Bayesian inference. Instead of point estimates, it produces full posterior distributions over the hazard ratios, naturally incorporating prior knowledge and providing coherent uncertainty quantification even with small samples or informative censoring. | Bayesian Survival Regression combines parametric or semiparametric survival models — such as Weibull, log-normal, or Cox proportional hazards — with Bayesian inference. Instead of point estimates, it produces full posterior distributions for regression coefficients and the baseline hazard, naturally handling censored observations and incorporating prior knowledge about event times or covariate effects. |
| ScholarGate数据集 ↗ |
|
|