In this project I explored medical insurance data in order to come up with a model to accurately predict the charges an insurance company can expect to pay for a given beneficiary’s healthcare based on payouts for previous beneficiaries, especially those that are in some way similar to the beneficiary in question. Given the need to importance of both accuracy and explainability in a setting like this, I focused on linear models throughout the report. The main deliverable here is the project report itself, but I've also included the code and data for reference.