Comparison of some methods to estimate the Cox proportional hazards model for interval-censored data
Interval censored data is common in several areas of knowledge, such as: epidemiology, finance, demo- graphy, medicine, among others. They occur when the event of interest, the failure time, is not observed exactly, but is within some interval of the observation time. Often in this situation an i...
| Main Authors: | , , |
|---|---|
| Format: | Online |
| Language: | spa |
| Published: |
Universidad Pedagógica y Tecnológica de Colombia
2022
|
| Subjects: | |
| Online Access: | https://revistas.uptc.edu.co/index.php/ciencia_en_desarrollo/article/view/12785 |
| _version_ | 1801706349972684800 |
|---|---|
| author | Bustos Giraldo, Olga Alexandra Jaramillo Elorza, Mario César Lopera Gómez, Carlos Mario |
| author_facet | Bustos Giraldo, Olga Alexandra Jaramillo Elorza, Mario César Lopera Gómez, Carlos Mario |
| author_sort | Bustos Giraldo, Olga Alexandra |
| collection | OJS |
| description |
Interval censored data is common in several areas of knowledge, such as: epidemiology, finance, demo- graphy, medicine, among others. They occur when the event of interest, the failure time, is not observed exactly, but is within some interval of the observation time. Often in this situation an imputation is made of the data that is not exactly known. Some methods of multiple imputation proposed in the literature are the PMDA (Poor Man’s Data Augmentation) algorithm and the ANDA (Asymptotic Normal Data Augmen- tation) algorithm, which allow estimating the parameters of the Cox proportional hazards model using classical estimation methods. There are also alternative methods to make these estimations such as the ICM (Iterative Convex Minorant) algorithm and a Bayesian approach, which do not impute the data with interval censoring.
In this work, a comparison was made via simulation of the performance of the estimators of the Cox model parameters produced by each of the aforementioned methods. The results showed that in general terms the ICM methods and the Bayesian approach present higher coverage probability values and lower mean square errors, in addition when increasing the sample size these values significantly improve compared to the PMDA and ANDA multiple imputation methods. In the latter, there were no significant differences between the results. Finally, an application was made with real data associated with a study of mastitis in milk cattle.
|
| format | Online |
| id | oai:oai.revistas.uptc.edu.co:article-12785 |
| institution | Revista Ciencia en Desarrollo |
| language | spa |
| publishDate | 2022 |
| publisher | Universidad Pedagógica y Tecnológica de Colombia |
| record_format | ojs |
| spelling | oai:oai.revistas.uptc.edu.co:article-127852023-06-26T20:42:54Z Comparison of some methods to estimate the Cox proportional hazards model for interval-censored data Comparación de algunos métodos para estimar el modelo de riesgos proporcionales de Cox para datos con censura a intervalo Bustos Giraldo, Olga Alexandra Jaramillo Elorza, Mario César Lopera Gómez, Carlos Mario Métodos de imputación múltiple, censura a intervalo, enfoque Bayesiano, algoritmo ICM (Iterative Convex Minorant), modelo de riesgos proporcionales de Cox Multiple imputation methods, Interval-censored, Bayesian approach, ICM (Iterative Convex Minorant) algorithm, Cox proportional hazards model. Interval censored data is common in several areas of knowledge, such as: epidemiology, finance, demo- graphy, medicine, among others. They occur when the event of interest, the failure time, is not observed exactly, but is within some interval of the observation time. Often in this situation an imputation is made of the data that is not exactly known. Some methods of multiple imputation proposed in the literature are the PMDA (Poor Man’s Data Augmentation) algorithm and the ANDA (Asymptotic Normal Data Augmen- tation) algorithm, which allow estimating the parameters of the Cox proportional hazards model using classical estimation methods. There are also alternative methods to make these estimations such as the ICM (Iterative Convex Minorant) algorithm and a Bayesian approach, which do not impute the data with interval censoring. In this work, a comparison was made via simulation of the performance of the estimators of the Cox model parameters produced by each of the aforementioned methods. The results showed that in general terms the ICM methods and the Bayesian approach present higher coverage probability values and lower mean square errors, in addition when increasing the sample size these values significantly improve compared to the PMDA and ANDA multiple imputation methods. In the latter, there were no significant differences between the results. Finally, an application was made with real data associated with a study of mastitis in milk cattle. Los datos con censura a intervalo son comunes en varias áreas del conocimiento, tales como: epidemiolo- gía, finanzas, demografía, medicina, entre otras. Ocurren cuando el evento de interés, el tiempo de falla, no se observa exactamente, sino que se encuentra dentro de algún intervalo del tiempo de observación. Con frecuencia en esta situación se realiza una imputación de los datos que no se conocen exactamente. Algunos de los métodos de imputación múltiple propuestos en la literatura son el algoritmo PMDA (Poor Man’s Data Augmentation) y el algoritmo ANDA (Asymptotic Normal Data Augmentation), los cuales per- miten estimar los parámetros del modelo de riesgos proporcionales de Cox utilizando métodos clásicos de estimación. También existen métodos alternativos para realizar estas estimaciones, como el algoritmo ICM (Iterative Convex Minorant) y un enfoque Bayesiano, que no realizan imputación de los datos con censura a intervalo. En este trabajo se realizó una comparación vía simulación del desempeño de los estimadores de los pa- rámetros del modelo de Cox producidos por cada uno de los métodos anteriormente mencionados. Los resultados evidenciaron que en términos generales los métodos ICM y el enfoque Bayesiano presentan va- lores de probabilidad de cobertura más altos y errores cuadráticos medios más bajos, además al aumentar el tamaño de la muestra estos valores mejoran notablemente comparados con los métodos PMDA y ANDA. En estos últimos no se evidenciaron diferencias considerables entre los resultados. Finalmente, se realizó una aplicación con datos reales asociados a un estudio de mastitis en ganado lechero. Universidad Pedagógica y Tecnológica de Colombia 2022-01-29 info:eu-repo/semantics/article info:eu-repo/semantics/publishedVersion application/pdf https://revistas.uptc.edu.co/index.php/ciencia_en_desarrollo/article/view/12785 10.19053/01217488.v13.n1.2022.12785 Ciencia En Desarrollo; Vol. 13 No. 1 (2022): Vol. 13 Núm. 1 (2022): Vol 13, Núm.1 (2022): Enero-Junio; 79-92 Ciencia en Desarrollo; Vol. 13 Núm. 1 (2022): Vol. 13 Núm. 1 (2022): Vol 13, Núm.1 (2022): Enero-Junio; 79-92 2462-7658 0121-7488 spa https://revistas.uptc.edu.co/index.php/ciencia_en_desarrollo/article/view/12785/12523 |
| spellingShingle | Métodos de imputación múltiple, censura a intervalo, enfoque Bayesiano, algoritmo ICM (Iterative Convex Minorant), modelo de riesgos proporcionales de Cox Multiple imputation methods, Interval-censored, Bayesian approach, ICM (Iterative Convex Minorant) algorithm, Cox proportional hazards model. Bustos Giraldo, Olga Alexandra Jaramillo Elorza, Mario César Lopera Gómez, Carlos Mario Comparison of some methods to estimate the Cox proportional hazards model for interval-censored data |
| title | Comparison of some methods to estimate the Cox proportional hazards model for interval-censored data |
| title_alt | Comparación de algunos métodos para estimar el modelo de riesgos proporcionales de Cox para datos con censura a intervalo |
| title_full | Comparison of some methods to estimate the Cox proportional hazards model for interval-censored data |
| title_fullStr | Comparison of some methods to estimate the Cox proportional hazards model for interval-censored data |
| title_full_unstemmed | Comparison of some methods to estimate the Cox proportional hazards model for interval-censored data |
| title_short | Comparison of some methods to estimate the Cox proportional hazards model for interval-censored data |
| title_sort | comparison of some methods to estimate the cox proportional hazards model for interval censored data |
| topic | Métodos de imputación múltiple, censura a intervalo, enfoque Bayesiano, algoritmo ICM (Iterative Convex Minorant), modelo de riesgos proporcionales de Cox Multiple imputation methods, Interval-censored, Bayesian approach, ICM (Iterative Convex Minorant) algorithm, Cox proportional hazards model. |
| topic_facet | Métodos de imputación múltiple, censura a intervalo, enfoque Bayesiano, algoritmo ICM (Iterative Convex Minorant), modelo de riesgos proporcionales de Cox Multiple imputation methods, Interval-censored, Bayesian approach, ICM (Iterative Convex Minorant) algorithm, Cox proportional hazards model. |
| url | https://revistas.uptc.edu.co/index.php/ciencia_en_desarrollo/article/view/12785 |
| work_keys_str_mv | AT bustosgiraldoolgaalexandra comparisonofsomemethodstoestimatethecoxproportionalhazardsmodelforintervalcensoreddata AT jaramilloelorzamariocesar comparisonofsomemethodstoestimatethecoxproportionalhazardsmodelforintervalcensoreddata AT loperagomezcarlosmario comparisonofsomemethodstoestimatethecoxproportionalhazardsmodelforintervalcensoreddata AT bustosgiraldoolgaalexandra comparaciondealgunosmetodosparaestimarelmodeloderiesgosproporcionalesdecoxparadatosconcensuraaintervalo AT jaramilloelorzamariocesar comparaciondealgunosmetodosparaestimarelmodeloderiesgosproporcionalesdecoxparadatosconcensuraaintervalo AT loperagomezcarlosmario comparaciondealgunosmetodosparaestimarelmodeloderiesgosproporcionalesdecoxparadatosconcensuraaintervalo |
