Quote:
Originally Posted by einbert
Oh **** off. If you want me to believe anything you have to say you're going to have to prove it.
I mean it should be pretty obvious to even the most casual observer the writers of the article are in no ways mathematicians. Yet despite this, they attempt to combine the numbers in two separate studies with some "look at me i can do division" level analysis and intentionally mislead the reader in to associating the results of their "analysis" with the legitimacy of Harvard or the Congressional Budget Office.
Even if we were to believe the legitimacy of the two underlying studies, the author divides an estimate of insurance rate on a national level with a correlation estimate to death rate on the state level. Take an estimate, divide by another estimate on partial data, add a pinch of appeal to authority protect yourself a bit by stating its not an exact science.
Might as well divine the stars.
I am also skeptical of the 2 underlying studies but without looking at the methodology, i cant comment. I will say that usually when looking at prediction models of highly complex systems for something as far away as 9 years, nobody knows.
Last edited by amoeba; 03-18-2017 at 12:10 PM.