Competing Hazards Regression Parameter Estimation Under Different Informative Priors

  • H Rehman Department of Statistics, Ramanujan School of Mathematical Sciences, Pondicherry University, Puducherry-605 014, India
  • Navin Chandra Department of Statistics, Ramanujan School of Mathematical Sciences, Pondicherry University, Puducherry-605 014, India
  • Fatemeh Sadat Hosseini-Baharanchi Minimally Invasive Surgery Research Center & Department of Biostatistics, School of Public Health, Iran University of Medical Sciences, Tehran, Iran
  • Ahmad Reza Baghestani Physiotherapy Research Center & Department of Biostatistics, Faculty of Paramedical Sciences, Shahid Beheshti University of Medical Sciences, Tehran, Iran;
Keywords: Competing risks, cause specific hazard, Cox regression, Burr type XII distribution, Bayes estimation, MCMC algorithm

Abstract

In the analysis of survival data, cause specific quantities of competing risks get considerable attention as compared to latent failure time approach. This article focuses on parametric regression analysis of survival data using cause specific hazard function with Burr type XII distribution as a baseline model. We obtained maximum likelihood and Bayes estimates of cumulative cause specific hazard functions under competing risk setup. For Bayesian point of view we proposed a class of informative priors for parameters to observe the comprehensive compatibility and their effectiveness under two different loss functions. The appropriateness of model is measured by the simulation study. Finally, we illustrate the proposed methodologies using bone marrow transplant data from the Princess Margaret Hospital Ontario, Canada.

Author Biographies

H Rehman, Department of Statistics, Ramanujan School of Mathematical Sciences, Pondicherry University, Puducherry-605 014, India

H. Rehman was born in 1993. He is a Ph.D. student in the Department of statistics, Pondicherry University, India, since August, 2017. He attended the Aligarh Muslim University, Aligarh, India where he received his B.Sc. (Hons.) in Statistics in 2014 and M.Sc. in Statistics in 2016. His research interests include survival analysis and its application, competing risks models, Bayesian estimation, and statistical computing.

Navin Chandra, Department of Statistics, Ramanujan School of Mathematical Sciences, Pondicherry University, Puducherry-605 014, India

N. Chandra serving as permanent faculty member in Department of Statistics, Ramanujan School of Mathematical Sciences at Pondicherry University, Pondicherry, India. He received Ph.D. Statistics from Banaras Hindu University, India and recipient of Senior Research Fellowship under major research project(MRP) sponsored by Department of Science and Technology, MHRD, Government of India. He supervised number of students for M.Sc. dissertation and research students for Ph.D. thesis in Statistics disciplines. He has completed MRP sponsored by University Grants Commission, India. His research interests includes Bayesian Inference, Reliability Theory (Estimation in accelerated life testing, Stress-Strength Models, Augmenting strength Models, bivariate models in reliability) and Survival Analysis (New life time distributions, characterizations and application, Estimation in Competing risks modelling, survival modelling of time to events in medical data).

Fatemeh Sadat Hosseini-Baharanchi, Minimally Invasive Surgery Research Center & Department of Biostatistics, School of Public Health, Iran University of Medical Sciences, Tehran, Iran

Fatemeh Sadat Hosseini-Baharanchi received her B.Sc. degree in Statistics and M.Sc. degree in Biostatistics from Tehran University of Medical Sciences; and Ph.D. degree in Biostatistics in 2016 from Tarbiat Modares University, Iran. Fatemeh was a visiting scholar in University of Connecticut, USA 2015. She is working in Iran University of Medical Sciences, Iran as an assistant professor in Biostatistics department since 2016. She is experienced in statistical modeling especially survival modeling and joint modeling of survival data and longitudinal measurements, especially in medical area. In addition, Fatemeh, as a member of international federation of Inventors Association, Geneva, Switzerland, helps students to pass the process of idea-to-patent as well as inventors to protect their intellectual property. She’s been focused on personal development, personal branding and innovative business models generation since 2018.

Ahmad Reza Baghestani, Physiotherapy Research Center & Department of Biostatistics, Faculty of Paramedical Sciences, Shahid Beheshti University of Medical Sciences, Tehran, Iran;

Ahmad Reza Baghestani received his B.Sc. degree in Statistics from Shahid Beheshti University and M.Sc. degree in Statistics from Tehran University of Medical Sciences; and Ph.D. degree in Biostatistics in 2010 from Tarbiat Modares University, Iran. He is associate professor in Biostatistics department in Shahid Beheshti University of Medical Sciences, Iran since 2011. He is fully-experienced in survival modeling, joint modeling, defective distributions, and competing risks analysis.

Published
2021-01-03
Section
Articles