Hacker News Reader

Comment by pc2g4d | Hacker News Reader

The answer is more complicated than that. Pearl has shown that it some situations it's possible to answer causal questions without running experiments. Basically, you can ask causal questions using the "do-calculus" he developed:

P(Dog barks | do(kick the dog) )

The `do` modifies your model to treat "kick the dog" as observed, breaking its dependence on any other variables, e.g. the doorbell ringing or the cat hissing. With those links broken, the model becomes simpler, and often the causal question can be answered using observational rather than experimental quantities.

At least, that's the idea. Not sure how broadly accepted it is, but Pearl is a towering figure and his work seemed robust at least to my not-so-expert eyes.

View on HN

That would still be a strong correlation, but not causation. One could add a weight to each observable state and select for the biggest weight so far, but it would have nothing to do with the actual causes. Would a dead dog bark when kicked?

There are some principles which cannot be undone by any amount of hipsterism and sophisticated sectarian bullshitting. There is no way to jump from observation to causality without knowing the implementation. At least in this particular universe.

In the realm of models (or ideas) it could be seem doable, but a map is not a territory, model does not represent reality until proven experimentally.

Well, I'd agree that you probably can't go from correlation to causation without introducing assumptions - but there are situations where you can try to make assumptions.

For example, in the social sciences it's common to look for "instrument variables" which are random and can't be affected by anything else. If you can find these types of variables then you can pretend that you're looking at the results from an experiment.

There are also methods like discontinuity analysis, which argue that if your system is discontinuously effected by a variable, but the causes will effect the variable in a continuous way, then you can look at changes around that specific point as being causal.

If the experiment you need to prove a model is an experiment that occurs naturally without intervention, you can prove a model just with observation.

An experiment must be carefully designed according to the working hypothesis and control. Natural phenomena is not a formal experiment.

Are you saying that it's impossible to draw conclusions from natural experiments? The 'family size' example here is nice: https://en.m.wikipedia.org/wiki/Natural_experiment#Recent_ex...

I don't see why you need to design a new experiment if data that satisfy your requirements already exist.

For example, if I wanted to know whether people given the name 'george' are more or less likely than the population as a whole to have at least one child, must I recruit pregnant women, and assign half of them to name their child 'george' and allow the others to choose any name? Would it not be better to look at existing data to compare Georges and non-Georges, perhaps slicing by socioeconomic variables? The results would be instant (rather than requiring 65 years to gather) and the dataset larger.

Here's the canonical example:

A is correlated to C. B is correlated to C. A is NOT correlated to B.

How is that possible?

The argument is that, if causation is unidirectional and acyclical, then the only causal structure that leads to the above is that A and B both cause C.

There's no other way to do it. And you can derive causation from observational data!

Of course in real life there are a bunch of problems with this, particularly measurement error and hidden unmeasured or unmeasurable variables. But more or less those exist with experimental data as well. You can only conclude things about observed variables and a very large number of unmeasured variables could throw off your inference.

Anyway, I more or less held your belief on this until I read Pearl's (and colleagues') material. There's much more expansion to it than the canonical example above.

Assuming unproven causality is exactly how uneducated primitive people used to form their religious beliefs.

Churches are related to people. People are related to the Sun. The Sun is caused by the prayers of the people in churches. Sun is in the sky today because someone somewhere prayed.

> There is no way to jump from observation to causality without knowing the implementation. At least in this particular universe.

At the extreme end of the spectrum, you have to weigh the probability that the statistical evidence is high enough to suggest causation against the probability that your mind isn't plugged into a simulation and all of the measurements have been faked. If the former is greater than the latter, than I'd say for all practical purposes we can call the phenomenon "causality".

Reply on news.ycombinator.com