The Jupyter Notebook attempts to use public data from The National Longitudinal Study of Adolescent to Adult Health (AddHealth) dataset to study the relationship between obesity and depression among adolescents collected from a stratified sample of 80 U.S high schools. Three additional waves of data have been collected and used to analyse the prevalence of any relationships as the adolescents move into adulthood. More details of the sampling design from Add Health can be found on the website in the references below. A report reviewing the correlations and findings has also been attached as a PDF.
This is part of an assignment from Imperial College to generate a prediction model using the Add Health dataset with health-related variables of personal interest and produce a report summarizing the case study, chosen models, prediction outcomes and future recommendations.
Add Health Dataset: https://addhealth.cpc.unc.edu/