Properties, estimation, and applications of the extended log-logistic distribution.
Sci Rep
; 14(1): 20967, 2024 Sep 09.
Article
in En
| MEDLINE
| ID: mdl-39251622
ABSTRACT
This paper presents the exponentiated alpha-power log-logistic (EAPLL) distribution, which extends the log-logistic distribution. The EAPLL distribution emphasizes its suitability for survival data modeling by providing analytical simplicity and accommodating both monotone and non-monotone failure rates. We derive some of its mathematical properties and test eight estimation methods using an extensive simulation study. To determine the best estimation approach, we rank mean estimates, mean square errors, and average absolute biases on a partial and overall ranking. Furthermore, we use the EAPLL distribution to examine three real-life survival data sets, demonstrating its superior performance over competing log-logistic distributions. This study adds vital insights to survival analysis methodology and provides a solid framework for modeling various survival data scenarios.
Full text:
1
Collection:
01-internacional
Database:
MEDLINE
Language:
En
Journal:
Sci Rep
Year:
2024
Document type:
Article
Affiliation country:
Kenia
Country of publication:
Reino Unido