« first day (3535 days earlier)      last day (1496 days later) » 
00:00 - 20:0020:00 - 00:00

8:11 PM
somebody with basic knowledge of relations?
 
8:27 PM
Wassup!
 
If anyone does stats here, please give this a look:
1
Q: Determining the power of the test in this question

ArcherThe following problem is from Devore's Probability and Statistics for Engineering and the Sciences, 8th edition, exercise 8.1 question 33: Reconsider the accompanying sample data on expense ratio(%) for large-cap growth mutual funds first introduced in Exercise 1.53: I cannot figure...

 
8:44 PM
Definition of $R(A)$ when $R$ is a relation and $A$ a set is $R(A)=\{y: xRy \space \text{for some} \space x \space \text{in} A\}$.

I have seen a theorem that $R(A \cap B) \subseteq R(A) \cap R(B)$, but the union "behaves nice" because of $R(A \cup B) = R(A) \cup R(B)$.

I would like to know why equality is not attained when also intersections are considered?

First, if $R$ is any (or almost any) relation, what are the conditions on the sets $A$ and $B$ so that equality is attained?

And second, if $A$ and $B$ are any (or almost any) sets, what are the minimal conditions on relations so tha
 
Hi, can anyone answer a super basic satistics question?
Basically, it is this
If I am asked for the expected value of a specific data point, that is E[$X_1$]
is this the same as if I am asked E[X]
?
 
Does it make sense to define the limit of a transfinite sequence of sets by

$$\forall x \left(x \in \lim_{\beta \rightarrow \alpha} f_\alpha \leftrightarrow \exists \beta (\beta < \alpha \land \forall \gamma (\beta \leq \gamma < \alpha \rightarrow x \in f_\gamma))\right)$$
 
9:02 PM
@Masterphile $R(A\cap B)=R(A)\cap R(B)$ for all $A,B$ is equivalent to $xRy\land x^{\prime}Ry\Rightarrow x=x^{\prime}$ for all $x,x^{\prime},y$ (this is well-known for $R$ a function)
 
any function?
what if intersection of A and B is empty?
 
@user76284 I would say it makes sense to define the $\liminf$ of that sequence as that
@Masterphile that's where the condition comes from, essentially
 
yes, but what if $R(A\cap B)=R(A)\cap R(B)$ only for some $A$ and $B$?
not necessarily for all
then $R$ is somewhere a function?
 
idk, the more general case appears to be uglier
 
yes, there are some "issues", basically i am trying to generalize and research basic theorems about relations
also for n-ary relations
i haven´t seen anywhere a comprehensive treatment of relations
they are just a chapter or section in set-theory or general-topology books
 
9:18 PM
-1
Q: $\Bbb R^2$ to torus?

geocalc33Consider some piecewise maps from $\Bbb R^2\to \Bbb R^2$ $$f(x_1, x_2) = (e^{x_1}, e^{x_2}).~~~~~x_1,x_2 \le 0$$ $$ g(x_1, x_2) = (e^{-x_1}, e^{-x_2})~~~~~~ x_1,x_2 \ge 0$$ $$ h(x_1,x_2)= (-e^{-x_1}, -e^{-x_2})~~~~~~x_1,x_2\ge0$$ $$ j(x_1, x_2) =(-e^{x_1}, -e^{x_2}). ~~~~~x_1,x_2 \le 0$$ Thes...

@EnjoysMath hi
 
@geocalc33 hey
I'm coding on BananaCats again
 
nice
 
the downvoters are haunting me also :P
 
It will understand some logic
 
I like downvotes
 
9:21 PM
yes, especially when you have a score of -3 and question gets closed, as it happened to me at least once :D
 
it's happened to me like a few times
 
@Thorgott I think this might be the intersection of what are usually considered the set-theoretic lim inf and lim sup.
 
yeah, i can try to phrase all future questions so to attract downvotes, i know a way
 
Modified for ordinal-indexed sequences. I have to check that, though.
 
I'm gonna try to get 10 downvotes today
I'll say I solved the (Rimean hyperthes) problem
 
9:25 PM
@Masterphile the thing is that this injectivity-like condition really only has to do with the equality when we consider $A,B$ as singletons, I don't think there's an easier characterization if you just look at arbitrary sets
 
nah, try with this: 1+1/2+...+1/n - ln(n) is irrational for every n, henceforth, the limit, euler-mascheroni constant is irrational as it is the limit of only irrational numbers @geocalc33
 
okay^
 
@user76284 what you defined is $\bigcup_{\beta<\alpha}\bigcap_{\beta\le\gamma<\alpha}f_{\gamma}$
 
@Thorgott i am not sure, i can think about that
 
which would be the analogon of the $\liminf$
 
9:28 PM
but hahahah do not write therefore but henceforth :D
@geocalc33
 
What kind of mathematical structure encodes idele classes? Is there some geometry I should be aware of?
How about that?
Henceforth :) I like that
 
or, was Banach actually Tarski because, paradoxically, he found a way to duplicate himself?
do you know about BTP?
 
Yeah
 
you can try also with this:

Since every irrational number can be approximated with rationals, and rationals are everywhere dense, is it true that every rational is dense in the $\mathbb R^n$, with the usual metric?
hardly you can obtain -10 or more
 
Okay
I'm gonna ask a good question
10 upvotes
 
9:35 PM
+10 is not so hard, but -10 is extremely hard
 
I already have +9
on two q's
maybe I can pull of a langrange multiplier as they say in france
 
mine record is +20 , on not so attractive question, also a basic question
that´s on this account (the only i have now), i had years ago one account with a question with, i think, about +60
 
is there a way to apply binomial formula on $\sum_{n=1}^{\infty}\binom{2n}{n}p^nq^n$
 
no
 
9:56 PM
How does such a sum even arise, @Simple?
 
random walk back to the origin after 2n steps with probability p step up and q steps down
 
@Khallil i had to consider yesterday $$\sum_{k=0}^{+ \infty} \sum_{r=0}^{k}{k \choose r} \dfrac {1}{(\sum_{w=1}^{+ \infty}a_w)^r}$$
 
I'm a bit confused as to how the ${2n}^C_{n}$ corresponds to getting back to the origin after $2n$ steps. It's been a while since I did any discrete stochastic, @Simple.
 
do not ask me where that arises
:D
elementary NT
that´s where
 
Well, you must have gone $n$ steps to the right and $n$ steps to the left if you return to the origin after $2n$ steps
 
10:08 PM
Wait steps to the left and right? I thought it was up and down :0
 
after a change of coordinates, we can WLOG assume the random walk takes place horizontally and in increments of 1.23 metric foot
 
LOL
Touché
 
10:33 PM
I am trying to show the sum converges with $p\neq q$ without using Stirling formula
 
11:04 PM
@Simple Isn't it the generalised binomial theorem applied to $(1-4x)^{-1/2}$, i.e. $$\frac{1}{\sqrt{1-4x}} = \sum_{n=0}^\infty \binom{2n}{n} x^n$$
The generating function for central binomial coefficients.
 
11:36 PM
@NoName I was thinking the binomial formula,
 
what can be said about open sets in $\mathbb R^n$?
 
they contain a neighborhood around each of their points
 
that´s almost the definition, i know that, tell me more
 
@Simple What do you mean?
 
they, uh, have positive Lebesgue measure
 
11:47 PM
yeah, since they are "fat"
what else?
 
What does 'fat' mean when you use it to describe a set, @Masterphile?
 
@Khallil homeomorphic to $\mathbb R^n$
 
that's not necessarily true
 
not if the set is empty, when else?
 
for example when your open set is disconnected
 
11:50 PM
@NoName I just couldn't recall this formula :)
 
ah yes, i was thinking of one-piece open sets
all the time
they have nice properties
the other ones also have, other from the fact they are not path-connected
 
even then, a punctured disk is not homeomorphic to the plane, for example
 
yes, but let us focus on one piece open-sets
 
if "one piece" means connected, the punctured disk is that
 
not on homeomorphims, i meant
they come later
okay, i accept, how would you map punctured disk onto the disk?
 
11:58 PM
why would I do that?
if I had to, I would say a surjection exists because they're equicardinal
 
00:00 - 20:0020:00 - 00:00

« first day (3535 days earlier)      last day (1496 days later) »