In this paper, we present and apply three alternative regression models (GLM) to the Nigerian Demographic and Health Survey (NDHS) 2018 data. The bivariate Poisson, the logistic and alternative binary models with extra dispersion parameter/s are also applied to the data which is a subset of the National Data. Our analyses focus primarily on the North-Central Zone States of Nigeria which include the Nigerian Federal Capital territory (FCT)-Abuja. We also employed the generalized linear mixed models (GLMM) to the selected data set, which has 21,656 observations. The response variable is the number of living children as at the time of the survey. The data has several covariates, but we have elected to use six of such covariates. Since the data is clustered (252 clusters in all) and nested within seven states, thus various sandwich estimators are considered. Three-level mixed effects models are also employed. SAS PROC NLMIXED, R packages and STATA modules are employed in all our analyses in this paper.

bivariate-Poisson, sandwich estimators, NDHS, under-dispersion, mixed models.

Received: December 1, 2021; Accepted: February 4, 2022; Published: March 3, 2022

How to cite this article: Bayo H. Lawal, Modeling the 2018 Nigerian demographic health survey data, Far East Journal of Applied Mathematics 112 (2022), 31-63. DOI: 10.17654/0972096022004

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

References:

[1] F. E. AlMuhayfith, A. A. Alzaid and M. A. Omair, On bivariate Poisson regression models, Journal of King Saud University - Science 28 (2016), 178-189.
[2] P. M. Altham, Two generalizations of the binomial distribution, Appl. Statist. 27 (1978), 162-167.
[3] P. Borges, J. Rodrigues, N. Balakrishnan and J. L. Bazán, A COM-Poisson type generalizations of the binomial distribution and its properties and applications, Statist. Probab. Lett. 87 (2014), 158-166.
[4] R. W. Conway and W. L. Maxwell, A queuing model with state dependent service rates, Journal of Industrial Engineering 12 (1961), 132-136.
[5] D. R. Cox, The analysis of multivariate binary data, Applied Statistics 21 (1972), 113-120.
[6] E. A. H. Elamir, Multiplicative-binomial distribution: some results and characterization, inference and random data generation, J. Stat. Theory Appl. 12(1) (2013), 92-105.
[7] F. Famoye, A new bivariate generalized Poisson distribution, Stat. Neerl. 64(1) (2010), 112-124.
[8] R. A. Francis, S. R. Geedipally, S. D. Guikema, S. S. Dhavala, D. Lord and S. Larocca, Characterizing the performance of the Conway-Maxwell-Poisson generalized linear model, Risk Analysis 32(1) (2012), 167-183.
[9] Ali Genc, The two-sided power-binomial distribution for modeling count data, Comm. Statist. Theory Methods 48(10) (2018), 2548-2561.
[10] C. Gourieroux, A. Monfort and A. Trognon, Pseudo maximum likelihood methods: applications to Poisson models, Econometrica 52 (1984), 701-720.
[11] R. C. Jang and R. Winkelmann, Two aspects of labor mobility: a bivariate Poisson regression approach, Empir. Econom. 18 (1993), 543-556.
[12] J. Kadane, G. Shmueli, G. Minka, T. Borle and P. Boatwright, Conjugate analysis of the Conway Maxwell Poisson distribution, Bayesian Anal. 1 (2006), 363-374.
[13] D. Karlis and I. Ntzoufras, Bivariate Poisson and diagonal inflated bivariate Poisson regression models in R, Journal of Statistical Software 14 (2005), 1-36.
[14] S. Kocherlakota and K. Kocherlakota, Bivariate Discrete Distributions, Marcel Dekker, New York, 1992.
[15] P. Kumaraswamy, A generalized probability density function for double-bounded random processes, Journal of Hydrology 46(1) (1980), 79-88.
[16] H. B. Lawal, Tables of percentage points of Pearson’s goodness-of-fit statistic for use with small expectations, Applied Statistics 29 (1980), 292-298.
[17] B. Lawal, Categorical Data Analysis with SAS and SPSS Applications, Taylor & Francis, New York, 2003.
[18] H. B. Lawal, On some mixture models for over-dispersed binary data, International Journal of Statistics and Probability 6(2) (2017a), 1-13.
doi: 10.5539/ijsp.v6np134.
[19] H. B. Lawal, GLM for some class of Com-Poisson distributions with applications, International Journal of Statistics and Probability 7(6) (2018), 1-13.
doi: 10.5539/ijsp.v7n6p.
[20] H. B. Lawal, Employing the double, multiplicative and the Com-Poisson binomial distributions for modeling over- and under-dispersed binary data, Journal of Advances in Mathematics & Computer Science 23(3) (2017b), 1-17.
doi: 10.9734/BJMCS/2017/33475.
[21] H. B. Lawal, Some new results on overdispersed models for the human sex ratio data, J. Biostatistics and Biom. Appls. 3 (2018a), 1-11.
[22] H. B. Lawal, On modeling the double and multiplicative binomial models as log-linear models, Advances in Computer Sciences 1(1) (2018b), 1-10.
doi: 10.31021/acs.20181103.
[23] J. K. Lindsey and P. K. Altham, Analysis of the human sex ratio by using overdispersion models, Appl. Statist. 47 (1998), 149-157.
[24] J. K. Lindsey and G. Mersch, Fitting and comparing probability distributions with log linear models, Comput. Statist. Data Anal. 13(4) (1992), 373-384.
[25] G. Lovison, An alternative representation of Altham’s multiplicative-binomial distribution, Statist. Probab. Lett. 36 (1998), 415-420.
[26] D. Lord, S. D. Guikema and S. R. Geedipally, Application of the Conway-Maxwell-Poisson generalized linear model for analyzing motor vehicle crashes, Accident Analysis and Prevention 40(3) (2008), 1123-1134.
[27] C. Manjor, P. Wijekoom and R. D. Yapa, The McDonald generalized beta-binomial distribution: a new binomial mixture distribution and simulation based comparison with its nested distributions in handling overdispersion, International Journal of Statistics and Probability 2(2) (2013), 24-41.
[28] National Population Commission (NPC) [Nigeria] and ICF Macro, Nigeria Demographic and Health Survey 2013, Abuja, Nigeria: National Population Commission and ICF Macro, 2018.
[29] J. C. Pinheiro and D. M. Bates, Approximations to the log-likelihood function in the non-linear mixed-effects model, J. Comput. Graph. Statist. 4 (1995), 12-35.
[30] K. F. Sellers and G. Shmueli, A flexible regression model for count data, The Annals of Applied Statistics 4 (2010), 943-961.
[31] K. F. Sellers, S. Borle and G. Shmueli, The COM-Poisson model for count data: a survey of methods and applications, Applied Stochastic Models in Business and Industry (2012), 104-116. doi: 10.1002/asmb.918.
[32] G. Shmueli, T. Minka, J. Borle and P. Boatwright, A useful distribution for fitting discrete data: revival of the Conway-Maxwell-Poisson distribution, J. R. Stat. Soc., Series C 54 (2005), 127-142.
[33] J. G. Skellam, A probability distribution derived from the binomial distribution by regarding the probability of success as a variable between sets of trials, J. R. Statist. Soc. B 10 (1948), 257-261.
[34] S. Kotz and J. R. van Dorp, Beyond Beta: Other Continuous Families of Distributions with Bounded Support and Applications, World Scientific, Singapore, 2004, pp. 308.