Related Experiment Videos
Bias-corrected maximum likelihood estimator of the negative binomial dispersion parameter
1Department of Mathematics and Statistics, University of Windsor, 401 Sunset Avenue, Windsor, Ontario N9B 3P4, Canada.
Biometrics
|March 2, 2005
Summary
A new bias-corrected maximum likelihood estimator for the negative binomial dispersion parameter offers improved accuracy and efficiency. This statistical method outperforms existing estimators in most comparative analyses.
Area of Science:
- Statistics
- Biostatistics
- Statistical Modeling
Background:
- The negative binomial distribution is frequently used to model count data.
- Accurate estimation of the dispersion parameter is crucial for valid statistical inference.
- Existing estimators for the negative binomial dispersion parameter have limitations in bias and efficiency.
Purpose of the Study:
- To derive a first-order bias-corrected maximum likelihood estimator for the negative binomial dispersion parameter.
- To compare the performance of the new estimator against existing methods.
Main Methods:
- Derivation of a first-order bias-corrected maximum likelihood estimator.
- Comparative analysis of bias and efficiency with other estimators, including maximum likelihood, moment, and extended quasi-likelihood methods.
- Application to a two-parameter negative binomial model and negative binomial regression.
Main Results:
- The proposed bias-corrected maximum likelihood estimator demonstrates superior bias and efficiency properties in most scenarios.
- The new estimator provides a more accurate and reliable estimation of the negative binomial dispersion parameter.
Conclusions:
- The first-order bias-corrected maximum likelihood estimator is a valuable improvement for analyzing negative binomial data.
- This enhanced estimator offers better statistical performance for dispersion parameter estimation.