3 ms·
> Not really no, which is why we have the whole correlation != causation thing. The number of people who drowned in a swimming pool perfectly correlates with po
by bko 5y ago
> Not really no, which is why we have the whole correlation != causation thing. The number of people who drowned in a swimming pool perfectly correlates with power generated by US nuclear plants over 10 years
Not every encounter with a body of water results in drowning. But nearly every drowning involves a body of water. It's not about correlation. It's bayesian probability.
P(drowning|water) >> P(drowning)
P(lung cancer|smoking) >> P(lung cancer)
P(inflation|money printing) >> P(inflation)
Do you get it yet?
> The same is true here - you need a model to adequately explain otherwise you're just intuiting, and intuition is frequently wrong.
Pretty much any econ 101 book. Search basic economics in Amazon for sources
- arcticbull 5y ago> Pretty much any econ 101 book. Search basic economics in Amazon for sources Only the Austrian ones, which are long since abandoned.
- arcticbull 5y ago> Do you get it yet? Every example you listed is an intuitive correlation. You can address the causative relationship between these, but not with the data you provided. All you've done is list a bunch of correlations, and asserted a causative relationship based on shared implicit context. There may be one, but you have not demonstrated it. Let's work an example without implicit context. I tell you: P(A|B) >> P(A) What have you learned about B->A? Literally nothing because the relationship between A and B could be any of: - A causes B (direct causation); - B causes A (reverse causation); - A and B are both caused by C (common causation); - A causes B and B causes A (bidirectional or cyclic causation); - There is no connection between A and B; the correlation is a coincidence.