Abstract:
|
The purpose of this project is to find distributions which best model body mass index (BMI) data. BMI has become a standard health indicator and numerous studies have been done to examine the distribution of BMI. Due to the skew and bimodal nature of BMI data, we focused on modeling with flexible skew distributions. We applied the models to UWEC BMI data and to empirical data as well. We used maximum likelihood estimation technique to obtain the models' parameters. Then we compared flexible models to more conventional distributions, such as skew-normal, and skew-t distributions using AIC and BIC. Our results indicate that the skew-t and alpha-skew Laplace distributions are able to describe unimodal BMI accurately whereas mixture of normal and finite mixture of scale mixture of skew normal distributions show better alternatives to both unimodal and bimodal conventional distributions. We believe the models discussed here will offer a framework for testing features such as - bimodality, asymmetry, and robustness- of the BMI data, thus providing a more detailed and accurate understanding of the distribution of BMI.
|