3 ms·
Everyone seems very focused on the (illuminating) distinction between median and mean. Are we maybe not thinking enough about there’s some level of difference
by benreesman 3y ago
Everyone seems very focused on the (illuminating) distinction between median and mean.
Are we maybe not thinking enough about there’s some level of difference between the two that’s a real issue?
- cscurmudgeon 3y agohttp://www.paulgraham.com/inequality.html http://www.paulgraham.com/inequality.html
- benreesman 3y agoA rare miss. pg doesn’t completely fuck up almost ever, I think that essay is tied with calling Altman a successful first-time founder for “even pg blows it sometimes”.
- cscurmudgeon 3y agoI disagree. This is a spot on article. If you celebrate diversity in humans and their activities and thoughts, then expect diversity in wealth. Irrational otherwise.
- benreesman 3y agoNo one (sane) is arguing for a Gini coefficient of zero. Some people are going to have more money. But the false dichotomy that Randian villains use is that it’s either that or fucking go wild. Vast wealth inequality is almost synonymous with capture. Rich people capturing the levers of society and then advantaging their dumbass kids and friends is like, what actually happens. The diversity poetry is a crinkly piece of tin foil everyone is supposed to look at while the profits get privatized and the losses get socialized. I’m all for capitalism if we can figure out how to do it without connections and shit being what really matters. But everyone who touched Loopt should have gone bankrupt, not like, take over the world.
- clipsy 3y agoI'd say plenty of people are thinking about wealth inequality, and the mean/median discrepancy is one of many relevant statistics.
- dasil003 3y agoAre you trying to say that an equitable wealth distribution would make mean and median across the population the same? This is a very specific statistical conclusion that is hard to grok without unpacking some of the assumptions behind it.
- benreesman 3y agoI’m saying that a healthy-ish Gini coefficient seems to be in the 0.2-0.275 type range, and a warlord won’t die in his bed above 0.85. It’s hard to get figures on us, but I used to get drunk with a bunch of Goldman quants who said we’ve been north of 0.5 once you cut the shit since 2008. It’s neither widely known nor considered polite to point out that the 5 ish mass shootings a day in the US make it per-capita more dangerous than Syria. So if you’re asking if I’m saying fuck rich people who exploit capture? Yeah: fuck them in particular.
- dasil003 3y agoI don’t disagree but it’s an obtuse way to make a point.
- benreesman 3y agoHaha fair play. HN is in general the best thing on the Internet but the number of people who stand with Greenspan, Rubin, and Summers on knee-capping Brooksley Born because they think the Ayn Rand car looks cool is like, very high. So I generally try to start that conversation in a roundabout way. I guess I’m ornery today. Objectivism is fascism clotted with purple prose written by a welfare recipient. Prove me wrong ;)
- jandrewrogers 3y agoThe distance between mean and median doesn’t tell you much. For any distance between the two, including them being equal, you can fit a great many very different distributions. They are indicative but the underlying distributions matter a lot.
- benreesman 3y agoYou are obviously correct in a general statistics sense. But I’m not sure that density functions give off more light than heat about this impending tennis court moment. In my city they have different tent camps for people who just got fired.