Far East Journal of Theoretical Statistics
Volume 67, Issue 3, Pages 233 - 242
(December 2023) http://dx.doi.org/10.17654/0972086323013 |
|
EFFECTS OF AGE-TRUNCATED FACTOR IN SURVIVAL ANALYSIS
John Darkwah
|
Abstract: There exists a class of alternative Cox models for analyzing time-to-event data that requires an appropriate adjustment of the age-truncated factor for better results. The adjustment is usually carried out by altering the factor to varying degrees in this class of models. This study seeks to assess the effects of this adjustment on parameter estimations, model significance, and the package of the R software used for the analysis. Univariate and multivariate models with the factor adjusted to varying degrees were fitted. The assessment of the effect on parameter estimation was carried out by considering variations in the parameter estimates and their respective standard errors, the statistical significance of the covariates and their proportional hazard statuses. The effect on the models was measured by considering their statistical and proportional hazard statuses. Finally, the effect on the package of the R software used for the analysis was assessed by considering its performance in the determination of the proportional hazard statuses of the models and the covariates. It was observed that the adjustment did not significantly affect parameter estimates, their standard errors, and the overall significance of the covariates. The same results were observed with the overall significance of the model. However, a constraint that hindered the performance of the package of the R software used for the analysis was observed when determining the PH status of some models and their respective covariates.
|
Keywords and phrases: age-truncated factor, proportional Hazard, alternative Cox models.
|
|
Number of Downloads: 47 | Number of Views: 178 |
|