Survival Analysis with Python

Survival analysis calculates time to failure using statistics, and Survival Analysis with Python provides a comprehensive guide to performing this analysis using the Python programming language. It covers parametric and non-parametric models, including maximum likelihood estimation, Kaplan–Meier estimator, and Cox-PH model, and demonstrates their application with practical examples and datasets.

\n Format: Hardback
\n Length: 84 pages
\n Publication date: 17 December 2021
\n Publisher: Taylor & Francis Ltd
\n

Survival analysis employs statistical techniques to estimate the duration until a specified event, such as failure, occurs. Survival Analysis with Python offers a comprehensive exploration of this intricate subject, demonstrating how to utilize the Python programming language to conduct survival analyses. Given the mathematical nature of survival analysis, which involves complex expressions and formulations, the book provides detailed explanations and examines practical implications. The book commences by providing a foundational overview of the concepts underlying statistical survival analysis. It subsequently delves into parametric models, covering topics such as the maximum likelihood estimate (MLE) of probability distribution parameters, the survival function, common probability distributions, and their analysis. The book also explores the exponential distribution as a survival function and the Weibull distribution as a survival function, presenting their derivations. Additionally, the book discusses non-parametric models, including the Kaplan–Meier (KM) estimator, derived using the MLE. An example dataset is fitted to the KM estimator, along with Python code and plotting curves. The Greenwoods formula and its derivation are also discussed. Models with covariates are explored, encompassing concepts such as time shift and the accelerated failure time (AFT) model, the Weibull-AFT model, the proportional hazard (PH) model, and the Cox-PH model. The significance of covariates is examined, and methods for selecting covariates are discussed. The Python lifelines library is utilized for coding examples throughout the book, making it a practical tutorial and valuable reference for practitioners.

\n Weight: 270g\n
Dimension: 175 x 329 x 12 (mm)\n
ISBN-13: 9781032148267\n \n