statistics

Heavy tails, scaling limits, R packages

Only ¹⁄₂ – ²⁄₃ of antidepressant users have depression. How do we use statistical learning to predict depression given antidepressant use?