Expected hours lost vs gained seems the most intuitive + rational model.
If thinking time was free, the appropriate theoretical way to handle uncertainty of course is to have multiple plausible distributions with different weights and multiply though the area-under-the-curve for each of them.
You probably want to include some discounting for future hours as well (since 1h now probably matters more than 2h 10 years from now), and another multiplier on the distribution for how much more vs less important your time after a large-scale catastrophe will be.
(Have I actually done this? No. But hopefully somebody else will).
Isn’t there a similar argument to covid – the best case scenario is bounded at zero hours lost, while the bound on the worst case is very high (losing tens of thousands of hours), so increasing uncertainty will tend to drag up the mean?
The current forecasts try to account for a bunch of uncertainty, but we should also add in model uncertainty – and model uncertainty seems like it could be really high (for the reasons in Dan’s comment). So this would suggest we should round up rather than down.
I was thinking altruistically, as was Ben. I agree that there are also the usual personal reasons to want to be alive, which pushes in favor of sacrificing expected work hours to increase the probability of survival.
Expected hours lost vs gained seems the most intuitive + rational model.
If thinking time was free, the appropriate theoretical way to handle uncertainty of course is to have multiple plausible distributions with different weights and multiply though the area-under-the-curve for each of them.
You probably want to include some discounting for future hours as well (since 1h now probably matters more than 2h 10 years from now), and another multiplier on the distribution for how much more vs less important your time after a large-scale catastrophe will be.
(Have I actually done this? No. But hopefully somebody else will).
Isn’t there a similar argument to covid – the best case scenario is bounded at zero hours lost, while the bound on the worst case is very high (losing tens of thousands of hours), so increasing uncertainty will tend to drag up the mean?
The current forecasts try to account for a bunch of uncertainty, but we should also add in model uncertainty – and model uncertainty seems like it could be really high (for the reasons in Dan’s comment). So this would suggest we should round up rather than down.
This seems wrong to me? It doesn’t take risk into account at all. I may be willing to pay some price to make sure dying is not a probable option.
In other words, plain expected value is just one metric. Conditional value at risk (with some risk tolerance) could be another, for example.
I was thinking altruistically, as was Ben. I agree that there are also the usual personal reasons to want to be alive, which pushes in favor of sacrificing expected work hours to increase the probability of survival.