A new adjusted Bayesian method in Cox regression model with covariate subject to measurement error

Hatice Işık; Duru Karasoy; Uğur Karabey

Makale detayları

A new adjusted Bayesian method in Cox regression model with covariate subject to measurement error

Yazarlar:

Hatice Işık | Duru Karasoy | Uğur Karabey

Yayıncı:

Hacettepe University

Ana Başlıklar:

Statistics

Dergi:

Hacettepe Journal of Mathematics and Statistics Vol.52 Issue 5, pp.1367-1378

Belge Tipi:

Makale

Yayın Tarihi:

10/2023

Belge Dili:

İngilizce

Document ID (Org No.):

2001301687

pISSN:

2651-477X

Anahtar kelimeler:

Özet:

 An important bias can occur when estimating coefficients by maximizing the known partial likelihood function in the Cox regression model with the measurement error covariate. We focus here on Bayesian methods in order to adjust measurement error and aim to propose an adjusting Bayesian method. Constructing simulation studies using Markov Chain Monte Carlo simulation techniques to investigate the performance of models. We compare the proposed method with the existing method that used partial likelihood function, Bayesian Cox regression model ignoring measurement error, the adjusted Bayesian Cox regression model that exists in the literature by a simulation study which consists of different sample sizes, censoring rates, reliability levels, and regression coefficients. Simulation studies indicate that the proposed method outperformed others given some scenarios. A real data set is analyzed for an illustration of the findings.

 An important bias can occur when estimating coefficients by maximizing the known partial likelihood function in the Cox regression model with the measurement error covariate. We focus here on Bayesian methods in order to adjust measurement error and aim to propose an adjusting Bayesian method. Constructing simulation studies using Markov Chain Monte Carlo simulation techniques to investigate the performance of models. We compare the proposed method with the existing method that used partial likelihood function, Bayesian Cox regression model ignoring measurement error, the adjusted Bayesian Cox regression model that exists in the literature by a simulation study which consists of different sample sizes, censoring rates, reliability levels, and regression coefficients. Simulation studies indicate that the proposed method outperformed others given some scenarios. A real data set is analyzed for an illustration of the findings.

 An important bias can occur when estimating coefficients by maximizing the known partial likelihood function in the Cox regression model with the measurement error covariate. We focus here on Bayesian methods in order to adjust measurement error and aim to propose an adjusting Bayesian method. Constructing simulation studies using Markov Chain Monte Carlo simulation techniques to investigate the performance of models. We compare the proposed method with the existing method that used partial likelihood function, Bayesian Cox regression model ignoring measurement error, the adjusted Bayesian Cox regression model that exists in the literature by a simulation study which consists of different sample sizes, censoring rates, reliability levels, and regression coefficients. Simulation studies indicate that the proposed method outperformed others given some scenarios. A real data set is analyzed for an illustration of the findings.

 An important bias can occur when estimating coefficients by maximizing the known partial likelihood function in the Cox regression model with the measurement error covariate. We focus here on Bayesian methods in order to adjust measurement error and aim to propose an adjusting Bayesian method. Constructing simulation studies using Markov Chain Monte Carlo simulation techniques to investigate the performance of models. We compare the proposed method with the existing method that used partial likelihood function, Bayesian Cox regression model ignoring measurement error, the adjusted Bayesian Cox regression model that exists in the literature by a simulation study which consists of different sample sizes, censoring rates, reliability levels, and regression coefficients. Simulation studies indicate that the proposed method outperformed others given some scenarios. A real data set is analyzed for an illustration of the findings.

Makale detayları

A new adjusted Bayesian method in Cox regression model with covariate subject to measurement error

Yazarlar:

Hatice Işık | Duru Karasoy | Uğur Karabey

Yayıncı:

Hacettepe University

Ana Başlıklar:

Statistics

Dergi:

Hacettepe Journal of Mathematics and Statistics Vol.52 Issue 5, pp.1367-1378

Belge Tipi:

Makale

Yayın Tarihi:

10/2023

Belge Dili:

İngilizce

Document ID (Org No.):

2001301687

pISSN:

2651-477X

Cox regression model

Bayesian analysis

polygonal function prior

measurement error

Sayı

Makale detayları

A new adjusted Bayesian method in Cox regression model with covariate subject to measurement error

Yazarlar: Hatice Işık | Duru Karasoy | Uğur Karabey

Yayıncı: Hacettepe University

Ana Başlıklar: Statistics

Dergi: Hacettepe Journal of Mathematics and Statistics Vol.52 Issue 5, pp.1367-1378

Belge Tipi: Makale

Yayın Tarihi: 10/2023

Belge Dili: İngilizce

Document ID (Org No.): 2001301687

pISSN: 2651-477X

Cox regression model

Bayesian analysis

polygonal function prior

measurement error

Sayı

Yazarlar:

Hatice Işık | Duru Karasoy | Uğur Karabey

Yayıncı:

Hacettepe University

Ana Başlıklar:

Statistics

Dergi:

Hacettepe Journal of Mathematics and Statistics Vol.52 Issue 5, pp.1367-1378

Belge Tipi:

Makale

Yayın Tarihi:

10/2023

Belge Dili:

İngilizce

Document ID (Org No.):

2001301687

pISSN:

2651-477X