Likelihood Approach for Bayesian Logistic Weighted Model
Missing Completely at Random Case
Abstract
Increasing the response rate and minimizing non-response rates represent the primary challenges to researchers in performing longitudinal and cohort research. This is most obvious in the area of paediatric medicine. When there are missing data, complete case analysis makes findings biased. Inverse Probability Weighting (IPW) is one of many available approaches for reducing the bias using a complete case analysis. Here, a complete case is weighted by probability inverse of complete cases. The data of this work is collected from the neonatal intensive care unit at Erbil maternity hospital for the years 2012 to 2017. In total, 570 babies (288 male and 282 females) were born very preterm. The aim of this paper is to use inverse probability weighting on the Bayesian logistic model developmental outcome. The Mental Development Index (MDI) approach is used for assessing the cognitive development of those born very preterm. Almost half of the information for the babies was missing, meaning that we do not know whether they have cognitive development issues or they have not. We obtained greater precision in results and standard deviation of parameter estimates which are less in the posterior weighted model in comparison with frequent analysis.
Downloads
References
D. Wolke, B. Sohne, B. Ohrt and K. Riegel. Follow-up of preterm children: Important to document dropouts. Lancet, vol. 345, no. 8947, p. 447, 1995.
S. Johnson, S. E. Seaton, B. N. Manktelow, L. K. Smith, D. Field, E. S. Draper, N. Marlow and E. M. Boyle. Telephone interviews and online questionnaires can be used to improve neurodevelopmental follow-up rates. BMC Research Notes, vol. 7, p. 219, 2014.
D. B. Rubin. Inference and missing data. Biometrika,vol. 63, no. 3, pp. 581-592, 1976.
M. Hofler, H. Pfister, R. Lieb and H. U. Wittchen. The use of weights to account for non-response and drop-out. Social Psychiatry and Psychiatric Epidemiology, vol. 40, no. 4, pp. 291-299, 2005.
L. Lazzeroni, N. Schenker and J. Taylor. Robustness of Multiple-imputation Techniques to Model Misspecification. United States: Proceedings of the Survey Research Methods Section, American Statistical Association, pp. 260-265, 1990.
B. L. Carlson and S. Williams. A Comparison of Two Methods to Adjust Weights for Non-response: Propensity Modeling and Weighting Class Adjustments. United States: Proceedings of the Annual Meeting of the American Statistical Association, 2001.
C. Agostinelli and L. Greco. Weighted Likelihood in Bayesian Inference. New York: Proceedings of the 46th Scientific Meeting of the Italian Statistical Society, 2012.
S. R. Seaman and I. R. White. Review of inverse probability weighting for dealing with missing data. Statistical Methods in Medical Research, vol. 22, no. 3, pp. 278-295, 2013.
S. R. Seaman, I. R. White, A. J. Copas and L. Li. Combining multiple imputation and inverse‐probability weighting. Biometrics, vol. 68, no. 1, pp. 129-137, 2012.
D. Clayton, D. Spiegelhalter, G. Dunn and A. Pickles. Analysis of longitudinal binary data from multiphase sampling. Journal of the Royal Statistical Society, vol. 60, no. 1, pp. 71-87, 1998.
R. J. Little and D. B. Rubin. Statistical Analysis with Missing Data. Chichester: Wiley, p. 5, 1987.
L. H. Curtis, B. G. Hammill, E. L. Eisenstein, J. M. Kramer and K. J. Anstrom. Using inverse probability-weighted estimators in comparative effectiveness analyses with observational databases. Medical Care, vol. 45, no. 10, pp. S103-S107, 2007.
Copyright (c) 2020 Dler H. Kadir
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
Authors who publish with this journal agree to the following terms:
1. Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License [CC BY-NC-ND 4.0] that allows others to share the work with an acknowledgment of the work's authorship and initial publication in this journal.
2. Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgment of its initial publication in this journal.
3. Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).