I would say that the benefits of leaving the house, once considering an arbitrary sequence of days, would far outweight any possible harm; simply because you would day of starvation in your nice little home, with 100% certainty (no risk involved here!). Unless there's people bringing you food - themselves risking devastating harm by doing so. And who says that you don't risk devastating harm in your home? Floods, fires, tornadoes, whatever. You might be saved by going OUT!
You argument and my response are very interesting, but keep in mind that they bear no link of any kind to the argument given in the paper. In the paper they're talking about risks at global level. For example they consider nuclear meltdowns (not nice stuff) as local risks. Bottom line: nobody cares about you or your house, or me or my house for that matter. Your comparison doesn't hold.
I do not consider the work proposed without flaws, but let's try to criticize it at the right level.