Hacker News new | past | comments | ask | show | jobs | submit login

This is junk analysis because there will always be signals that randomly correlate. There are a lot of signals out there that we can plot pairwise and the axises here aren't identical. All this says is that long term interest rates formed a hump shape, and crime formed a hump shape. There are so many hump shapes it is just a matter of finding 2 humps that peaked around the same time.

Not only that, but in this case peak interest rates were around the infamous 1970s era when a lot of policies change. So even if these are related, there is probably a hidden variable that a lot of policies changed at the same time and the two series still aren't causally related. There was likely a policy refresh on how crime gets handled at the same time as there was a policy reset on how interest rates were managed.




Yeh, this. The frequency of babies named Jeffrey[1] follows this pattern. Maybe all this crime was committed by Jefferys?

[1] https://namerology.com/2022/03/08/the-ultimate-baby-name-gra...


Yeah, this is exactly what I thought. Interest Rates correlate to a lot of things. Seems like stretch to say there is a cause, not just correlation.

Like, there is more crime in cities, but, because there are also more people. What is per-capita crime. Small town with 10 people, 1 felon, man, they are 10% felons.


> Seems like stretch to say there is a cause, not just correlation.

A better take might be that they reliably happen together.




Consider applying for YC's W25 batch! Applications are open till Nov 12.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: