Hacker News Reader

Comment by jey | Hacker News Reader

You can infer a lot about the underlying causal processes from the "pile of correlations". Humans do this all the time. If flipping a light switch correlates highly with the lamp's state switching from "on" to "off" and vice-versa, and in every observed instance the light-switch was flipped before the light changed state, wouldn't you assign a higher probability to the causal model "flipping the switch causes lamp to change state" than "lamp changing state causes switch to flip"? In other words, we can do a lot better than to just say that "lamp changing state and switch-flipping are highly correlated".

http://plato.stanford.edu/entries/causation-probabilistic/

(Sorry for being pedantic.)

View on HN

In this case we can be reasonably confident in the causal link between the switch and the light because we're controlling the switch ourselves -- this helps us eliminate the other causal possibilities (a) that the light going on and off causes the switch to flip and (b) both the light and the flipping of the switch are separately caused by some third factor of which we're ignorant.

In situations where we can't freely vary any of the parameters we're always going to have a lot more difficulty. Given a pile of correlations between, say, happiness and the countless other variables in the mere two hundred or so countries which exist (eg "average bovine thigh circumference"), it'd be impossible, in the absence of any good theories about what should make people happy, to determine what does make people happy.

Luckily we have pretty good ideas from our own observations and from those of others about what actually does make people happy: health, wealth, nice weather, absence of civil war, et cetera. But we'll never be able to get anything other than the vaguest confirmation of what we already believed out of statistical methods alone.

Or, we could just go back to the original topic of the article...people in Costa Rica seem to be happy.

Disbanding their military (which I was shocked to read, honestly) and spending the money (presumably) on education, amazingly, seemed to result in a better country? Hell, most countriesd if they disbanded their military and burned the money instead would be better off.

It seems to me, assuming this survey is remotely correct, is people enjoy a life where they are not sticking their noes into other countires business, and having nice weather and a strong social construct is pleasurable.

I can make the same inferences even if I'm just observing another person or some robot flipping the switch, so no, it doesn't depend on us "controlling" the switch.

But the robot _might_ be being controlled by some guy in another room who is also controlling the lightbulb (alternatively the person could be following instructions given by an earpiece). Or maybe they're both hooked up to a geiger counter. I admit it seems pretty unlikely that anybody would wire a robot or person up to flip a switch at exactly the same time a lightbulb comes on, but it only seems unlikely because of our pre-existing knowledge about robots and lightbulbs. If we take away that knowledge and simply make it a statement about how variable A changes when variable B does, it's much harder to infer which way, if either, the causal link goes.

(I suppose if you really wanted to be paranoid you could suppose that your _own_ actions in flipping the lightswitch might be subconsciously being controlled by someone else, but this is a whole different level of skepticism)

Reply on news.ycombinator.com