|
|||||
|
|
Firth adjustment for Weibull current-status survival analysis |
|
| Authors: | Hung-Mo Lin John M. Williamson Hae-Young Kim |
| Affiliation: | 1. Department of Population Health Science and Policy, Icahn School of Medicine at Mount Sinai, New York, NY, USA;2. Hung-Mo.Lin@mountsinai.org;4. Division of Parasitic Diseases and Malaria, National Center of Global Health, Centers for Disease Control and Prevention (MS A-06), Atlanta, GA, USA;5. Department of Public Health, New York Medical College, Valhalla, NY, USA |
| Abstract: | AbstractAnalysis of right-censored data is problematic due to infinite maximum likelihood estimates (MLE) and potentially biased estimates, especially for small numbers of events. Analyzing current-status data is especially troublesome because of the extreme loss of precision due to large failure intervals. We extend Firth’s method for regular parametric problems to current-status modeling with the Weibull distribution. Firth advocated a bias reduction method for MLE by systematically correcting the score equation. An advantage is that it is still applicable when the MLE does not exist. We present simulation studies and two illustrative analyses involving RFM mice lung tumor data. |
| Keywords: | Bias correction current-status data parametric survival analysis small sample Weibull distribution |