good pedantry https://t.co/DwZBl7P4rk
-
-
Replying to @eigenrobot
I just can't grok this yet, I ran into it in a textbook and it still confuses the hell out of me
3 replies 0 retweets 5 likes -
Replying to @selentelechia
so just imagine a uniform distribution between eg 0 and 1 you make one draw from that distribution Say you get 0.728502174 The probability of getting a value in a given range is the area under the pdf for that range Area under a curve at a point when the curve is finite is 0
1 reply 1 retweet 8 likes -
Replying to @eigenrobot @selentelechia
So, the probability of any particular draw is zero under a continuous distribution, but conditional on pdf(X) > 0 it's still possible
4 replies 0 retweets 5 likes
What's really fun is like Imagine a distribution where pdf(x) = 0.2 for x in [0, 0.5) and pdf(x) = 0.8 for x in [0.5, 1] p(X=0.1) = p(X=0.8) = 0 BUT defined around a limit p(X=0.8) / p(X=0.1) = 4 ie in a certain sense 0.8 is still four times more likely that 0.1
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.