Life expectancy at birth is one of the crucial markers for assessing the populace’s well-being and healthcare arrangements of a nation. Like in other countries, the life expectancy of Bangladeshi people is increasing steadily. Be that as it may, it is to observe variability in life expectancy across all the districts in Bangladesh. Therefore, selecting a statistical model that captures the regional disparity is essential to get a valid statistical conclusion. To our knowledge, no endeavor is made so far to demonstrate the life expectancy data of Bangladesh that accounts for the district-wise variation. We perform a likelihood-based cluster analysis to investigate this circumstance in this paper. We employ a non-parametric mixture technique for the unobserved random effect in the modeling. Various model diagnostics plots of the residuals further reveal that the final model fits the life expectancy data of Bangladesh satisfactorily. Our findings also identify that for both urban and rural areas of the districts of Thakurgaon, Rangamati, and Panchagarh as having the lowest level of life expectancy growth. From further analysis, the results show that the urban areas of Bangladesh have significantly higher life expectancy than the rural areas.

life expectancy, likelihood based cluster analysis, random intercept modeling, bootstrap paired t-test, Bangladesh.

Received: November 17, 2022; Accepted: January 3, 2023; Published: February 6, 2023

How to cite this article: Tarikul Islam and Noor Jahan Akter, Statistical modeling of life expectancy data to identify the low life expectancy regions in Bangladesh: a likelihood based cluster analysis, Universal Journal of Mathematics and Mathematical Sciences 18(2) (2023), 165-184. http://dx.doi.org/10.17654/2277141723010

This Open Access Article is Licensed under Creative Commons Attribution 4.0 International License

References:

[1] A. F. Rabbi, Imbalance in life table: effect of infant mortality on lower life expectancy at birth, Journal of Scientific Research 5(3) (2013), 479-488.
[2] Bangladesh Bureau of Statistics (2020), Report on Sample Vital Registration System 2020.
[3] D. Böhning, S. Karasek, C. Terschüren, R. Annu and R. Fehr, A nonparametric random coefficient approach for life expectancy growth using a hierarchical mixture likelihood model with application to regional data from North Rhine-Westphalia (Germany), BMC Medical Research Methodology 13(1) (2013), 1-13.
[4] A. P. Dempster, N. M. Laird and D. B. Rubin, Maximum likelihood from incomplete data via the EM algorithm, Journal of the Royal Statistical Society: Series B (Methodological) 39(1) (1977), 1-22.
[5] G. V. Glass, P. D. Peckham and J. R. Sanders, Consequences of failure to meet assumptions underlying the fixed effects analyses of variance and covariance, Review of Educational Research 42(3) (1972), 237-288.
[6] M. M. Islam and K. M. A. K. Azad, Rural-urban migration and child survival in urban Bangladesh: are the urban migrants and poor disadvantaged? Journal of Biosocial Science 40(1) (2008), 83-96.
[7] M. S. Islam, M. I. Tareque, M. N. I. Mondal, A. M. Fazle Rabbi, H. T. Khan and S. Begum, Urban-rural differences in disability-free life expectancy in Bangladesh using the 2010 HIES data, PloS One 12(7) (2017), e0179987.
[8] F. Konietschke and M. Pauly, Bootstrapping and permuting paired t-test type statistics, Statistics and Computing 24(3) (2014), 283-296.
[9] J. Kuha, AIC and BIC: comparisons of assumptions and performance, Sociological Methods and Research 33(2) (2004), 188-229.
[10] B. G. Lindsay, The geometry of mixture likelihoods: a general theory, The Annals of Statistics 11 (1983), 86-94.
[11] Geoffrey J McLachlan and Thriyambakam Krishnan, The EM Algorithm and Extensions, John Wiley and Sons, Volume 382, 2007.
[12] D. A. V. I. D. Peel and G. MacLahlan, Finite Mixture Models, John and Sons, 2000.