Body Density Prediction with MLR
Development and Evaluation of a Multiple Linear Regression-Based Model for Predicting Body Density
Introduction
As per the World Health Organization, obesity is a significant risk factor for several chronic ailments, such as cardiovascular diseases and cancer. To quantify obesity, body fat mass (BFM) is a viable option. However, the accurate methods of measuring BFM are prohibitively expensive. Inexpensive alternatives, such as waist circumference and waist-to-hip ratio, have been shown to be related to BFM. In this study, we had access to a dataset containing BFM measurements (or, more precisely, body density measurements) along with other factors like height, weight, and age for males. Our primary objective was to develop and validate a regression model for estimating BFM. This involved various steps, such as residual analysis, handling outliers, analyzing multicollinearity, and selecting variables using different performance criteria.
The code and data sets used in our analysis can be found in this Github repo.
Demo
Below is an interactive demo of our final model. Feel free to try it out!