This work uses data aggregated from a number of sources including the American Community Survey census.gov, clinicaltrials.gov, and cancer.gov 🏥. The dataset contains 3048 datapoints and 33 variables.

🚩 In this project we build and compared several models to predict the "TARGET_deathRate". We tested ordinary least square (OLS) regression, CART and Random Forest algorithms. We also used VSURF for variable selection.

Models were compared using the root mean squared error (RMSE):

⚪ We condlude that the best model to predict "TARGET_deathRate" is Random Forest followed by step method forward or backward.