4 ms·
> This means that on average, we’ll need a fifty million times larger sample for the sample average to be as close to the true average as in the case of dice ro
by Dylan16807 3y ago
> This means that on average, we’ll need a fifty million times larger sample for the sample average to be as close to the true average as in the case of dice rolls.
This is "as close" in an absolute sense, right?
If I take into account that the lottery value is 20x larger, and I'm targeting relative accuracy, then I need 2.5 million times as many samples?
- dustypotato 3y agoDidn't make sense to me as well. Maybe it's better to use something along the lines of variance/mean instead?