This is repo contains the code and data for my posts [1, 2] about the returns to risk taking in science, philanthropy, and public policy. This is a follows on my earlier writing about the efficient frontier of philanthropy.
The conclusion is that there do seem to be returns to uncertainty, but not to downside risk. See the jupyter notebooks [1, 2] or blog posts [1, 2] for more details.