Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

Only one problem - your anecdote provides (very weak) evidence against Warren's conclusions.

You had a major medical cause but you didn't go bankrupt. Lots of people do. To properly estimate the number of medical bankruptcies, you need to do the following calculation:

    # Medical bankruptcies = # of bankruptcies x [P(bankruptcy | medical cause) - P(bankruptcy | no medical cause)] x P(medical cause)
All Warren measured was P(medical cause | bankruptcy). Then she led a bunch of innumerate reporters to believe that P(medical cause | bankruptcy) x # of bankruptcies = # of medical bankruptcies.

Your anecdote cuts against Warren's conclusions [1] because it suggests we increase our estimate of P(no bankruptcy | medical cause) and thereby reduce our estimate of P(bankruptcy | medical cause).

[1] I'm not suggesting a single anecdote proves much of anything. I'm just pointing out the direction it's infinitesimal value points.



Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: