Article: Survival analysis: a comparative study of frequentist and Bayesian approaches Journal: International Journal of Biometrics (IJBM) 2025 Vol.17 No.6 pp.570 - 595 Abstract: This experiment compares two categories of survival analysis, including traditional (Kaplan-Meier, Cox PH, parametric models, Bayesian Cox) and deep learning (DeepSurv, DeepHit), on 299 heart failure patients. DeepHit did better than the others (C-index = 0.75, IBS = 0.16) and it surpassed DeepSurv (0.73) and Cox PH (0.71). Cox PH turned out to be quicker and easily interpretable and age (HR = 1.05) and serum creatinine (HR = 1.37) emerged as key predictors. The Bayesian models performed well in terms of small-sample (DIC = 1,272.30), thus providing uncertainty quantification. Parametric models (e.g., Weibull, AIC = 1,282.24) were effective, where distributional assumptions were met. Important variables, such as age, ejection fraction, and renal biomarkers, were also always significant. The model selection is need-based: Cox PH is fast and easy to interpret; Bayesian procedures have small sample sizes and priors; DeepHit is used when the patterns are complex; and parametric models are when the data is distributed in a certain way. Inderscience Publishers - linking academia, business and industry through research

Title: Survival analysis: a comparative study of frequentist and Bayesian approaches

Authors: Zaheer Aslam; Abid Hussain; Nasir Ali; Muhammad Hanif; Roquia Aslam

Addresses: Department of Statistics, Government College Asghar Mall, Rawalpindi, Pakistan ' Department of Statistics, PMAS-Arid Agriculture University, Rawalpindi, Pakistan ' Department of Statistics, PMAS-Arid Agriculture University, Rawalpindi, Pakistan ' Department of Statistics, PMAS-Arid Agriculture University, Rawalpindi, Pakistan ' Combined Military Hospital, Muzaffarabad, AJK, Pakistan

Abstract: This experiment compares two categories of survival analysis, including traditional (Kaplan-Meier, Cox PH, parametric models, Bayesian Cox) and deep learning (DeepSurv, DeepHit), on 299 heart failure patients. DeepHit did better than the others (C-index = 0.75, IBS = 0.16) and it surpassed DeepSurv (0.73) and Cox PH (0.71). Cox PH turned out to be quicker and easily interpretable and age (HR = 1.05) and serum creatinine (HR = 1.37) emerged as key predictors. The Bayesian models performed well in terms of small-sample (DIC = 1,272.30), thus providing uncertainty quantification. Parametric models (e.g., Weibull, AIC = 1,282.24) were effective, where distributional assumptions were met. Important variables, such as age, ejection fraction, and renal biomarkers, were also always significant. The model selection is need-based: Cox PH is fast and easy to interpret; Bayesian procedures have small sample sizes and priors; DeepHit is used when the patterns are complex; and parametric models are when the data is distributed in a certain way.

Keywords: survival analysis; parametric models; non-parametric models; semi-parametric methods; Bayesian parametric models; Bayesian semi-parametric models.

DOI: 10.1504/IJBM.2025.149673

International Journal of Biometrics, 2025 Vol.17 No.6, pp.570 - 595

Received: 26 Feb 2025
Accepted: 16 May 2025
Published online: 10 Nov 2025 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article

Title: Survival analysis: a comparative study of frequentist and Bayesian approaches

Keep up-to-date