Advancing lifetime data modeling via the Marshall-Olkin cosine Topp-Leone distribution family | ||
| Computational Journal of Mathematical and Statistical Sciences | ||
| Article 5, Volume 4, Issue 2, November 2025, Pages 476-499 PDF (464.99 K) | ||
| Document Type: Original Article | ||
| DOI: 10.21608/cjmss.2025.374786.1155 | ||
| Authors | ||
| Abdulhameed Ado Osi* 1; Gaber Sallam Salem Abdalla2; Narinderjit Singh Sawaran Singh3; Ahmad Abubakar Suleiman1 | ||
| 1Department of Statistics, Aliko Dangote University of Science and Technology, Wudil 713101, Nigeria | ||
| 2Department of Insurance and Risk Management, Faculty of Business, Imam Mohammad Ibn Saud Islamic University (IMSIU), Riyadh 11432, Saudi Arabia | ||
| 3Faculty of Data Science and Information Technology, INTI International University, Persiaran Perdana BBN Putra Nilai 71800, Malaysia | ||
| Abstract | ||
| Probability distributions are fundamental tools in statistical modeling, particularly in the analysis of lifetime, reliability, and epidemiological data. Classical distributions such as the exponential, Weibull, and gamma, while analytically convenient, often lack the flexibility required to model complex real-world phenomena, such as skewness, heavy tails, and intricate dependence structures. In response to these limitations, this paper introduces a novel trigonometric-based extension of the Marshall-Olkin family, termed the Marshall-Olkin Cosine Topp-Leone (MOCTL) distribution family. This new family incorporates additional shape parameters that allow for greater modeling flexibility and adaptability across various data types. We derive and explore several important statistical properties of the proposed family, including its density, distribution, hazard rate, and quantile functions. Parameter estimation is addressed using the maximum likelihood estimation (MLE) method, and a detailed Monte Carlo simulation is conducted to assess the performance, bias, and consistency of the MLEs. The real-world applicability of the MOCTL family is demonstrated through three datasets, including medical and epidemiological studies. Furthermore, a log-MOCTL Weibull regression model is proposed and applied to HIV/TB and COVID-19 datasets, confirming its superior modeling capability. | ||
| Keywords | ||
| Marsharl-Olkin family; Cosine G family; Simulation study; Maximum likelihood estimation; Survival Regression Model; Cox-Snail residual Analysis | ||
|
Statistics Article View: 178 PDF Download: 138 |
||
