Comparison of conditional likelihood and modified score function approaches in the analysis of matched case-control data

Authors

  • Abdul Baten Department of Statistics, Jagannath University, Dhaka-1100,
  • Taslim S Mallick Department of Statistics, Biostatistics & Informatics, University of Dhaka, Dhaka-1000
  • Jafar A Khan Department of Statistics, Biostatistics & Informatics, University of Dhaka, Dhaka-1000

DOI:

https://doi.org/10.3329/bjsr.v26i1-2.20233

Keywords:

Conditional logistic regression, matched case-control data, maximum conditional likelihood, modified score function

Abstract

In this study, we compare Maximum Conditional Likelihood or MCL and Modified Score Function or MDS approach for modeling Antenatal Care (ANC) seeking behavior in matched case-control data obtained from Bangladesh Demographic and Health Survey 2007. The estimates of parameters are almost identical under both approaches. Due to the computational complexity of the MDS approach, one may prefer using the alternative computationally simpler MCL approach for the analysis of similar datasets.

DOI: http://dx.doi.org/10.3329/bjsr.v26i1-2.20233

Bangladesh J. Sci. Res. 26(1-2): 75-82, December-2013

Downloads

Download data is not yet available.
Abstract
858
PDF
814

Downloads

Published

2014-08-27

How to Cite

Baten, A., Mallick, T. S., & Khan, J. A. (2014). Comparison of conditional likelihood and modified score function approaches in the analysis of matched case-control data. Bangladesh Journal of Scientific Research, 26(1-2), 75–82. https://doi.org/10.3329/bjsr.v26i1-2.20233

Issue

Section

Articles