4 ms·
It's their point, and it's a good one, but I think they're somewhat overstating how common power-law data is; it probably varies a lot by field of study. And at
by nerdponx 3y ago
It's their point, and it's a good one, but I think they're somewhat overstating how common power-law data is; it probably varies a lot by field of study. And at least the logarithm of a power-law variable can help bring it back closer to the world of sanity. Plus, there are plenty of fields where nonparametric tests of medians are accepted standard practice.
- mjburgess 3y agoYou can turn most issues into powerlaws by recursing a reasonable risk distribution over it. So suppose we ask, what is our confidence in X? (rather than X); and then, what is our confidence in the model by which we give confidences in X (ie., the model risk); and so on... In practice, what we want to model is the appropriate confidence, not an actual prediction (bunk). So we are very often screwed. Statistics is an illusion.