In logistic regression, odd-ratio and its standard error computations are crucial. Some conventional approaches are available. Here, a simple variation is demonstrated, which leads to smaller standard errors and hence shorter confidence intervals. Bootstrap methods are also implemented and comparisons are made using several popular data sets.

logistic regression, confidence intervals.

How to cite this article: Mezbahur Rahman, A note on odds ratio standard error in logistic regression, Far East Journal of Theoretical Statistics 63(2) (2021), 85-91. DOI: 10.17654/0972086321002

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

References:

[1] D. W. Hosmer and S. Lemeshow, Applied Logistic Regression, Second ed., Wiley, New York, 2000.
[2] M. Rahman, K. A. Holt and P. A. O’Connor, Inferences in logistic models along with estimation of parameters using discriminant function approach, Far East Journal of Theoretical Statistics 39(1) (2012), 53-66.