Analyzing Postpartum Amenorrhea Duration: Estimating Weibull Distribution Parameters with EM Algorithm Using Current Status Data
,
Abstract
Introduction: The duration of post-partum amenorrhea (PPA), a crucial aspect of reproductive health, re- mains a significant factor in family planning and maternal well-being. Understanding the distribution of this period provides valuable insights into fertility patterns and informs contraceptive strategies. However, this duration often involves current status data, presenting challenges in accurate estimation and analysis.
This study aims to employ statistical modeling, specifically utilizing the Weibull distribution and the EM algorithm, to estimate parameters related to PPA duration. The primary objective is to develop a robust methodology for parameter estimation within current status data.
Methods: The research employs the Weibull distribution, known for its applicability in current status data analyses, as a framework for modeling PPA duration. Leveraging the EM algorithm, the study develops an approach to estimate the Weibull distribution parameters from the current status data. This methodology focuses on overcoming the challenges posed by interval-censored observations, providing a more accurate understanding of the duration.
Results: The application of the EM algorithm to estimate Weibull distribution parameters yields promising results. The methodology successfully addresses the complexities of current status data, offering estimates that enhance the
understanding of PPA duration. The results highlight the efficacy of the proposed approach in handling such nuanced datasets.
Conclusion: This study underscores the significance of statistical modeling techniques, particularly the Weibull distribution coupled with the EM algorithm, in estimating parameters for PPA duration analysis. The successful application of this methodology emphasizes its potential for furthering the understanding of fertility patterns and aiding in informed decision making concerning reproductive health strategies.
2. Tika Ram Aryal. Weibull distribution to describe the patterns of the duration of post-partum amenorrhea. 31(1), 2011.
3. TR Aryal. Retrospective reporting of the duration of post-partum amenorrhea: a survival analysis. Kath- mandu University medical journal, 4(2):211–217, 2006.
4. Elodie Brunel and Fabienne Comte. Cumulative distribution function estimation under interval censoring case 1. 2009.
5. Arthur P Dempster, Nan M Laird, and Donald B Rubin. Maximum likelihood from incomplete data via the em algorithm. Journal of the royal statistical society: series B (methodological), 39(1):1–22, 1977.
6. David B Dunson and Donna D Baird. A flexible parametric model for combining current status and age at first diagnosis data. Biometrics, 57(2):396–403, 2001.
7. Sam Efromovich. Efficient nonparametric estimation of distribution for current status censoring. J BIOSTAT EPIDEMIOL. VOL. 11, NO. 4, 2025 450 Kumar S et al. Analyzing Postpartum Amenorrhea Duration: Estimating Weibull Distribution Parameters with EM ...Electronic Journal of Statistics, 16(1):998–1057, 2022.
8. Tsai-Hung Fan, Yi-Fu Wang, and She-Kai Ju. A competing risks model with multiply censored reliability data under multivariate weibull distributions. IEEE Transactions on Reliability, 68(2):462–475, 2019.
9. Luís Andrade FerreirA and José Luís SiLvA. Parameter estimation for weibull distribution with right cen- sored data using em algorithm. Eksploatacja i Niezawodnos´c´, 19(2):310–315, 2017.
10. Kathleen Ford and Young Kim. Distributions of postpartum amenorrhea: some new evidence. Demography, pages 413–430, 1987.
11. Antonio Eduardo Gomes and Cibele Queiroz DA-SILVA. Nonparametric maximum likelihood estimation for current status data with misclassification. Rev. Bras. Biom, 29(1):102–120, 2011.
12. Piet Groeneboom and Kim Hendrickx. The nonparametric bootstrap for the current status model. 2017.
13. International Institute for Population Sciences (IIPS) and ICF.National Family Health Survey
(NFHS-5). https://dhsprogram.com/publications/ publication-fr347-dhs-final-reports.cfm, 2020. Accessed: [date accessed].
14. Nicholas P Jewell and Mark van der Laan. Current status data: Review, recent developments and open problems. Handbook of statistics, 23:625–642, 2003.
15. Mimi Y Kim, Victor G De Gruttola, and Stephen W Lagakos. Analyzing doubly censored data with covari- ates, with application to aids. Biometrics, pages 13–22, 1993.
16. Tamalika Koley and Anup Dewanji. Current status data with two competing risks and missing failure types: a parametric approach. Journal of Applied Statistics, 49(7):1769–1783, 2022.
17. MD Krailo and MC Pike. Estimation of the distribution of age at natural menopause from prevalence data.American journal of epidemiology, 117(3):356–361, 1983.
18. Anup Kumar, Abhishek Bharti, and RC Yadava. An adjustment of truncation and selection effect for estimating conception rate from first birth interval data. Journal of Data Science, 17(4):756 767, 2019.
19. Anup Kumar, Sachin Kumar, Jai Kishun, Uttam Singh, and Pushpraj Pushpraj. Estimation of the distribution of duration of breastfeeding from cross-sectional data: Some methodological issue. Journal of Biostatistics and Epidemiology, 6(4):290–304, 2020.
20. Thomas A Louis. Finding the observed information matrix when using the em algorithm. Journal
J BIOSTAT EPIDEMIOL. VOL. 11, NO. 4, 2025 451Kumar S et al. Analyzing Postpartum Amenorrhea Duration: Estimating Weibull Distribution Parameters with EM ...of the Royal Statistical Society Series B: Statistical Methodology, 44(2):226–233, 1982.
21. Ruchi Mishra, Kaushalendra K Singh, and P Singh Brijesh. Duration of post-partum amenorrhoea: A model- based approach. Indian Journal of Population and Development, 1(1):41–50, 2021.
22. Dilip C Nath, Atanu Bhattacharjee, and HB Singh. An estimation of PPA duration for manipuri women: A bayesian approach in zero-inflated longitudinal data analysis.
23. MS Panwar, Vikas Barnwal, and CP Yadav. A latent variable approach for modeling recall-based time-to- event data with weibull distribution. Computational Statistics, pages 1–32, 2024.
24. Sedigheh Mirzaei Salehabadi, Debasis Sengupta, and Rituparna Das. Parametric estimation of menarcheal age distribution based on recall data. Scandinavian Journal of Statistics, 42(1):290 305, 2015.
25. Girish Singh. Estimation of the duration of postpartum amenorrhea in the presence of some censored data.The Journal of Obstetrics and Gynecology of India, 57(1):53–57, 2007.
26. Pragya Singh, Kaushalendra K Singh, and Brijesh P Singh. An EM algorithm approach to estimate parame- ters of fluctuating nature of PPA. Indian Journal Population and Development, 1(2):285 294, 2021.
27. SN Singh, BN Bhattacharya, and RC Yadava. An adjustment of a selection bias in postpartum amenorrhea from follow-up studies. Journal of the American Statistical Association, 74(368):916920, 1979.
28. Jianguo Sun and John D. Kalbfleisch. The analysis of current status data on point processes. Journal of the American Statistical Association, 88(424):1449–1454, 1993.
29. Zhibin Tan. A new approach to mle of weibull distribution with interval data. Reliability Engineering & System Safety, 94(2):394–403, 2009.
30. Bert van Es and Catharina Elisabeth Graafland. Nonparametric kernel density estimation for univariate curent status data. arXiv preprint arXiv:1707.00544, 2017.
31. Kumar, S., Kumar, A., Misra, A. K., Kishun, J., & Singh, U. (2024). Adjustment of Truncation Effect in First Birth Interval using Current Status Data Technique. Journal of Biostatistics and Epidemiology.
32. Qian, C., Srivastava, D. K., Pan, J., Hudson, M. M., & Rai, S. N. (2024). Estimating transition
intensity rate on interval-censored data using semi-parametric with EM algorithm approach. Communications in Statistics-Theory and Methods, 53(17), 6038-6054.
| Files | ||
| Issue | Vol 11 No 4 (2025) | |
| Section | Original Article(s) | |
| DOI | https://doi.org/10.18502/jbe.v11i4.22520 | |
| Keywords | ||
| EM algorithm Weibull Distribution Current Status Data PPA | ||
| Rights and permissions | |
|
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. |
