China
Africa
Explore chapters and articles related to this topic
Introduction
Published in Catherine Legrand, Advanced Survival Models, 2021
We refer the reader to [186], and in particular Chapter 3, for a thoughtful discussion on the different mechanisms of censoring and truncation. As we will detail below, the presence of censoring and truncation has consequences on the building of the likelihood function and we will have to carefully account for the information available from each type of observation. Indeed, censored observation will not contribute in the same way as fully observed cases. Due to the presence of censoring, classical analysis techniques for continuous variables can therefore not be used. If present, truncation should ideally also be taken into accounted in the building of the likelihood function. Most of the time, only right-censoring is taken into account when analyzing the data, and a major part of the survival data analysis literature focus on right censoring. Note that the different right-censoring mechanisms actually lead to the same survival likelihood function, and can therefore be handled by the same survival analysis techniques [107]. This book will mainly focus on right-censored data, but whenever available, we will provide references for extension to other censoring and truncation schemes of the methods presented .
Survival Modeling I: Models for Exchangeable Observations
Published in Gary L. Rosner, Purushottam W. Laud, Wesley O. Johnson, Bayesian Thinking in Biostatistics, 2021
Gary L. Rosner, Purushottam W. Laud, Wesley O. Johnson
We care about censored data, because we want to include all information in the data set. Even censored observations contain information; a patient still alive 5 years after entering the study tells us something about the treatment, particularly if most patients with this disease tend to die within a year of diagnosis.
Survival Analysis
Published in Atanu Bhattacharjee, Bayesian Approaches in Oncology Using R and OpenBUGS, 2020
The survival data having censored data can be estimated by Kaplan-Meier estimator. It is a nonparametric estimator. The survival curves are compared by the Kaplan-Meier estimator. It is obtained as a product limit formula. The Kaplan-Meier estimator for the survival distribution function is estimated as
Regression modelling of interval censored data based on the adaptive ridge procedure
Published in Journal of Applied Statistics, 2022
Olivier Bouaziz, Eva Lauridsen, Grégory Nuel
Interval-censored data arise in situations where the event of interest is only known to have occurred between two observation times. These types of data are commonly encountered when the patients are intermittently followed up at medical examinations. This is the case for instance in AIDS studies, when HIV infection onset is determined by periodic testing, or in oncology where the time-to-tumour progression is assessed by measuring the tumour size at periodic testing. Dental data are another examples which are usually interval-censored because the teeth status of the patients are only examined at visits to the dentist. While interval-censored data are ubiquitous in medical applications, it is still a common practice to replace the observation times with their midpoints or endpoints and to consider these data as exact. This allows one to analyse the data using standard survival approaches but may result in a large bias of the estimators. In the present paper, we develop a new method for the analysis of time to ankylosis complication on a dataset of replanted teeth. The three main goals of our method are to adequately take into account interval-censoring, to be able to identify time ranges where patients are particularly at high risk of developing the complication and to investigate if a sub-population of non-susceptible patients exists.
Semiparametric inference for the scale-mixture of normal partial linear regression model with censored data
Published in Journal of Applied Statistics, 2022
Mehrdad Naderi, Elham Mirfarah, Matthew Bernhardt, Ding-Geng Chen
Censored data is widely seen in practical studies, such as time-to-death in cancer research, time-to-infection in HIV, TB and COVID-19 studies. The interval-censoring scheme is the most general scenario that covers the typical right- and left-censoring as special cases. Interval-censored data can also be generated from other science fields, such as detection limits of quantification in environmental, toxicological and pharmacological studies. We assume that the set of joint variables
Inference on Nadarajah–Haghighi distribution with constant stress partially accelerated life tests under progressive type-II censoring
Published in Journal of Applied Statistics, 2022
Sanku Dey, Liang Wang, Mazen Nassar
Censoring is a very common phenomenon in life testing experiments and reliability studies. Generally speaking, censoring implies that exact failure times are known for only a portion of the units under study where the units are lost or removed from test before failure due to time and cost limitations, and the data observed from such experiments are called censored data. There are many censoring schemes (CSs) in life tests, and the most common ones are Type-I and Type-II censoring (see, e.g. Lawless [27]), where the associated lifetime test is terminated after a fixed time or when exact units fail. Furthermore, due to rapid advancement in technology and increasing global competition, there is a constant stress on manufacturers to produce high-quality products. Life testing and reliability experiments often need to reduce the total testing time and cost. Therefore, a more general and flexible CS called the progressive Type-II censoring scheme was introduced in practice. In this CS, consider that n units are placed on a life test and n−1 surviving units. When the second failure occurs, mth failure, then all the remaining 16], Singh et al. [38], Lee and Cho [28] and the references therein. One can refer to the extensive review paper of Balakrishnan [5] as well as the monograph of Balakrishnan and Aggarwala [6].