Healthcare Data Analytics – Predictive Modeling of Medical Insurance Costs
Developed a comprehensive analysis of 1,338 policyholder records to identify key drivers of medical insurance charges. Conducted exploratory data analysis, statistical testing, and multiple linear regression, revealing that smoking status, BMI, and age are the strongest predictors of costs. Quantified findings smokers incur 280% higher charges and translated results in to actionable recommendations,including targeted wellness initiatives, age-specific preventive programs, and risk-based premium adjustments. Achieved an Adjusted R² of 0.7494, demonstrating strong model reliability for guiding pricing and risk management strategies.