« first day (4477 days earlier)      last day (826 days later) » 

user76284
user76284
03:59
0
Q: $k$-medians of common probability distributions

user76284$\DeclareMathOperator*{\argmin}{argmin} \DeclareMathOperator*{\expect}{E}$ Let $(X,d)$ be a metric space. Let $\mu$ be a probability measure on $X$. Its median is $$\argmin_{m \in X} \expect_{z \sim \mu} d(z, m)$$ Let its $k$-median be $$\argmin_{M \in [X]^k} \expect_{z \sim \mu} \inf_{m \in M} d...

references median robust central-tendency m-estimation
Has anyone seen this in the literature before? Perhaps under a different name?
usεr11852
usεr11852
04:47
Sometimes I think Granger causality is the Stats whipping boy...
 
8 hours later…
CowperKettle
CowperKettle
13:08
> Treatment refractory depression participants reported longstanding, unremitting, or recurrent depression with an average of 2.45 episodes (s.d. = 7.75)
I forgot. Can a standard deviation be larger than an average?
Kodiologist
Kodiologist
13:31
@gung-ReinstateMonica Okay, I didn't realize that was policy. I think the edit would've been perfectly okay if it had been correct, but it was wrong.
@CowperKettle For sure. The standard normal distribution has mean 0 and SD 1.
 
1 hour later…
CowperKettle
CowperKettle
14:43
@Kodiologist Oh. Thank you!
It has been too long a time.
I tend to forget everything.
 
2 hours later…
User1865345
User1865345
17:00
So far what I have observed re links being broken and dead: lecture notes, slides, working versions of papers and some sites. What remains obvious is the most reliable ones are the wiki links.

« first day (4477 days earlier)      last day (826 days later) »