« first day (1280 days earlier)      last day (4035 days later) » 

00:00
When you give your full voice to the axiom of choice, that's recursion.
(Because well-orderedness lies behind recursion, and sets can't be well-ordered without the choice axiom.)
When you explain something deep using terms somewhat crude, that's recursion.


When you fly through the sky, the simple terms, they won't lie, that's recursion.
When you give your full voice to the axiom of choice, that's recursion.
@robjohn are u there?
00:18
@Danny yes
a question, if u have time to just confirm
hold on let me write
Well said, Paul, well said - thanks once again
if i have an interval $[0,1]$, with $k$ partions which are equally long, with distribution $ Re(0,1)$. If now $X_1$ is the number of times that one or more observations occured in the first interval. what is the distribution ....on $X_1$
@robjohn i gave it $X_1 \sim Bin(n,1/k)$
do u agree
i forgot to write that there were $n$ observations made....
hi, anyone know the difference between the 1999 publication and the 2010 publication of Neukirch's Algebraic number Theory?
@Danny I guess it depends on what you mean by Bin(n,1/k)
@Danny you give a lot of terms that might mean something in a given context, but without context, I am not sure what you mean
00:34
@robjohn thats all i got maybe i got something wrong in translation
f
why isnt latex working
now, well the binomial distribution...
with $n$ total observations what is the probability that $k$ times "are good"
@Danny what do you mean by Re(0,1)? uniform density on (0,1)?
@robjohn yes
with $1/k$ i just mean the probability of getting in any interval with the uniform distribution
@Danny and by Bin(n,1/k) ? There are usually 2 integer values determining a binomial distribution and it doesn't sound as if you are asking for a binomial distribution
Is $X_1$ a random variable, or a distribution function?
Wait, I think I might be able to piece things together...
Do you mean something like $\binom{n}{j}(1/k)^j(1-1/k)^{n-j}$
@robjohn $X_1$ is a random variable X_1
@robjohn yes
That is the probability of getting exactly $j$ things in any given slot
00:49
that is the binomial distribution $Bin(n,1/k)$
yes
Then yes, you are right
but there is another thing i was wondering
iam a bit puzzled about this one since it seems so easy that it might not be right
if i define another random variable $Y_1 = \frac{X_1}{n}$ then what distribution does it have
i'd say it has the same
because $P(X_1)= P(X_1/n$)
when $X_1$ occurs then $X_1/n$ occurs
@robjohn
I'm having a bit of a struggle with some set-theory question, I manage to get half-way through but the last part eludes me... any set-theorists around?
@Danny $P(X_1)$ means nothing. You need to give a value, such as $P(X_1\gt4)$
ok yes
00:58
well, what do you mean?
look at the new varible defined $Y_1= \frac{x_1}{n}$
If you say that $P(X_1\gt1)=P(X_1/n\gt1)$, I would say no
iam asking for the distribution
That is what I would think you are asking from what you wrote
In that case, then $P(X_1)\ne P(X_1/n)$
whatever that really means
well what i mean is
$P(X_1 = k) = P(Y_1 = k)$
@robjohn dont u agree
01:04
@Danny When you write $X_1/n$ it means the values of the random variable $X_1$ divided by $n$. The density of $(X_1/n)n$ is the same as the density of $X_1$ but that is trivial
$P(X_1=k)=P(X_1/n=k/n)$
the random variable is defined as $Y_1 = X_1/n$
the new one
Yes, but then I wouldn't say that the distribution of $Y_1$ is the same as that for $X_1$ it is a scaled version of it
but that might be what you are thinking of
what iam going to do is to calculate the expectation value, varians
for $Y_1$
What are the axioms of set theory that we use today?
@robjohn but i would need the distribution to do that
01:15
Yes, and $\mathrm{E}(X_1)=n\mathrm{E}(Y_1)$. If they were the same, their expected values would be the same
@robjohn
@Danny This is why I say that $P(X_1)\ne P(Y_1)$. They give different expectation values
@saadtaame yes?
@robjohn okay so to get to the distribution of $Y_1$
Did you see my question? It's up there
how can i get it
01:18
@saadtaame It is not in my scrollback.. what is the question?
What are the axioms of set theory that we use today? @robjohn
@Danny First you have to state what you are asking properly. $P(X_1=k)=P(Y_1=k/n)$
@saadtaame It depends on what flavor of set theory you are considering. There are several flavors of set theory. You'd need to talk to a set theorist to get a good overview
Both $1$ and $-1$ are generators of the integers correct?
under addition
@Alex yes
can you prove it?
@Danny or equivalently, $P(Y_1=k)=P(X_1=nk)$
since $Y_1=X_1/n$
When proving it, how would one justify that you can subtract $1$'s to generate $0$ and the negative integers? That is the one thing I am a bit questionable on. Similarly for $-1$ and generating the positive half of the integers.
01:26
@robjohn so now i need to get the distribution out of this ...
Do you just acknowledge that $-1$ and $1$ are inverses of each other and state that one generates the other?
@TedShifrin
Will just go ahead and ask.. So, I have these:
f:R->R, where f(x)=-(3^-x), A={n+1-1/(k+1) : n,k natural numbers}, and B=f[A]. I have to prove that f is the isomorphic function from A to B, and that it is the only one. The first part is easy, I have no clue, however, how to go for the second part. I tried by stating that it is true and get to a contradiction, however I can't figure out how to get to one...
the isomorphic function in this case has to be onto, and keep the order. They don't mention any order, so I assume (hopefully rightfully assume) that in this case it's the regular order
01:42
@robjohn so what is $P(X_1 = nk)$
n and k are integers
@Pedro I'm about to nap
Sup?
Damn it Mike.
Is it quiiiick?
01:44
That's no help
Let $|G|=p^\alpha m, p\not\mid m$; $P$ a $p$-Sylow subgroup, and $N$ a normal subgroup.
Suppose $|N|=p^\beta n,p\not\mid n$.
You mean $|G|$ right
Okie.
It's \nmid
Or \not\mid
OK, keep going
I may fall asleep within seconds
I have to show $|P\cap N|=p^\beta$ and $|PN/N|=p^{\alpha-\beta}$-
Now, it is easy to see $|P\cap N|=p^\gamma$ with $\gamma\leqslant \beta$.
So I have to show $\gamma \geqslant \beta$.
01:48
Blurg.
I'll be no use to you right now.
I just sorta stared at that for a second and said "Sylow" and then closed my eyes
LOL. Sleep tight.
@robjohn like if $P(Y_1 = s ) = P(X_1 = ns)$ then it wouldnt work in the binomial distribution having $P(X_1 = ns) = \binom{n}{ns}\frac{1}{k}^{ns} (1-\frac{1}{k}^{n-ns})$
@Danny If you pay attention to the possible values of $s$ it does
@robjohn $s$ is an integer less than or equal to $n$
@PedroTamaroff yessir
02:01
@AlexanderGruber My question is just some lines above.
@PedroTamaroff here?
Yes! Finally got it!
@AlexanderGruber Yes.
well you want to prove that $P$ is a $p$-Sylow subgroup of $N$
$N\cap P$.
Yes.
@anon 'ello.
02:07
did you look at the orders of stuff with the (2nd?) iso theorem?
(the PN/N=P/P\cap N one)
hello
@PedroTamaroff so $P\subseteq PN\subseteq G$, what happens with the indices?
wlog one can assume P contains N's p-Sylow (hint: N is conjugation-invariant)
@robjohn i have no clue
02:09
@anon wlog because N is conjugation invariant?
mm hmm
for instance $|P\cap N|=|(P\cap N)^\sigma|=|P^\sigma\cap N^\sigma|=|P^\sigma\cap N|$, so just apply $\sigma$ to get $P$ lying over $N$'s $p$-sylow
also, N's p-Sylow is unique, yes?
@PedroTamaroff not necessarily
err, a $p$-sylow of $N$
whatevs
(consider $N=G$)
02:11
Right, duh.
what can you say about $[PN:P]$?
I have to show that is $p^{\alpha-\beta}$ =P
heh, alex already answered himself above
Ram
Ram
v(n) denotes the index of n relative to a fixed primitive root g, that is the exponent v for which g^v \equiv n.

I can't understand this, can some body explain me this definition. I mean g^v \equiv n wrt to what
with respect to the modulus
the modulus is not stated within the text you've quoted
Ram
Ram
02:15
@anon you mean mod q?
@robjohn ok i got it now $s $ is a rational
@Ram it doesn't matter what letter one uses. perhaps whatever source you're quoting from has chose its own letter
Ram
Ram
yes, its not stated clearly. Sorry, he assumes g is primitive root of mod a prime. I missed this
@anon, I forgot, the other day I made a stupid point by claiming, R^n+1 is universal cover of RP^n. If you remember.
@anon, and if you remember that discussion about, an open set in RP^n is there any surjective continuous map to R^n+1, can't we use some thing like peano curves?
@anon I couldn't do it.
03:01
howdy @Pedro, @anon, @Alex
03:14
@TedShifrin Heya.
I'm trying to prove Hölder Jordan.
Well, the first part.
That any nontrivial group admits a composition series.
I have to do it by induction on $|G|$.
And I guess I can assume $G$ is not simple.
In particular $|G|$ is not prime.
Nevermind, Jacobson has a proof and it is all pretty easy.
03:40
hey there
Hey, @Alex.
@TedShifrin have you been to calhoun, GA?
Nope, @Alex. Am I missing anything?
@Danny Sorry, I had to leave. Yes, if $X_i$ are integers, then $X_i/n$ are rational
Hey there
04:02
howdy folks
@Pedro: Mhenni is at it again. Agh.
Hi again @Jessy.
Hi Joe.
@TedShifrin i'm not sure yet. i'm meeting someone there next weekend (it's the midpoint between my old uni and this one) so i'm looking for things to do
Really a midpoint?
@TedShifrin under the highway topology? :)
Hi @TedShifrin
04:17
Better to go to Chattanooga ... Only a few more miles.
Hi @Jasper. @Alex: Dalton, Rome, Calhoun are all boring.
I need to lose 10 kg, lol.
Me too.
@TedShifrin atlanta's just a few more miles towards me, too. hmm.
I think it must be the lack of exercise that is responsible for my being overweight.
Exercise is good for all sorts of reasons, @Jasper.
yup, @Alex.
04:26
@TedShifrin Exercises in a math book are good too, lol.
@JasperLoy get a pull-up bar.
@AlexanderGruber I only do walking, lol.
@JasperLoy that's because you don't have a pull-up bar yet.
it makes you better at math. gets the blood flowin'
Sorry I think I asked this before maybe, but if I am doing a math grad program, should I learn French or German? Russian is out for me, lol.
@JasperLoy what type of math are you into?
04:28
@AlexanderGruber Well, nothing in particular. I guess I can only decide after the first year grad program...
well, i think french is probably the best choice in a general sense. german might be better for some disciplines (e.g. mine).
@AlexanderGruber And what discipline is that?
finite group theory.
Ah, I hate groups, lol.
i get that a lot.
04:32
@AlexanderGruber I actually prefer reading German translated books than French translated ones, so I guess maybe I should go for German.
@JasperLoy it's also better if you're a fan of heavy metal.
@AlexanderGruber Erm, why? lol
because many of the best heavy metal bands are german. also, i don't think french metal exists.
Also, I like German movies more than French ones.
I am installing TeX Live 2013, I burned my own DVD.
Hello, @Gruber, @Loy
@AlexanderGruber I much like fields.
I more field theorist than ring or group.
@Jasper
04:44
@BalarkaSen i'm learning ring theory lately in my algebra course
i've done some field theory but nothing too high minded just your average graduate galois theory
Well, if it's gotta ring, it's gotta ring, but if you have to ring, try power series rings. That's a heck of a thing.
@BalarkaSen i like the hilbert's basis theorem proof for power series
The main site says "This site is currently in read-only mode; we’ll return with full functionality soon." ??
What happened?
@BalarkaSen somebody probly spilled coffee on the servers over at SE.
Ah, it's okay now.
@AlexanderGruber =D
They say LHFs are profitable, no?
04:48
i am not sure what those are
@AlexanderGruber Yes.
@AlexanderGruber Low-Hanging-Fruits.
Actually, galois theory over power series is what I like. Strange beasts they are.
@BalarkaSen have you ever looked at profinite galois theory?
@AlexanderGruber Well, a little of here a little of there. I am onto topological galois theory lately.
@BalarkaSen is that different from the Krull topology?
@AlexanderGruber Not really.
@AlexanderGruber I am not sure what 'average graduate galois theory' is. I am not a graduate.
@AlexanderGruber Do you like commutative algebras?
05:02
so, I need to maximize the areas of intersect circled in blue here sdrv.ms/LJOW7u (both, not individually). I can move the diagonal rectangle how I like, but not the 2 parallel rectangles. Must I just brute-force it, or is there a more elegant method?
05:28
@BalarkaSen it's alright.
i'm just learning it. my professor isn't very good, i'm trying not to hold it against the discipline.
i like category theory.
@AlexanderGruber Meh.
my main complaint about C.A. and A.G. is that i don't really understand the motivation for most of it. i feel like i need a lot more examples, to see the objects it studies in action, before I can really appreciate the abstract theorems about them.
Someone in the forum bragged about his <10^5 listings of Pell solutions and I listed 10^146 for him.
Oh, 10^153 actually.
@BalarkaSen What's the point of that...?
05:53
@PedroTamaroff Nothing. Simply kicks.
@AlexanderGruber Have you tried inverse galois theory?
06:21
HI @BalarkaSen!
How to show if the rows are in proportion or equvalently columns are in proportion of matrix P then X and Y are independent random variables here?
@robjohn
Hey @Sush.
Doing Econ well?
07:19
some people who come here ask math questions and then veer off, become chatty and uninterested
@PedroTamaroff Does dividend has the same meaning as in banking?
The same meaning in math as in banking
 
1 hour later…
08:37
Hellllooo
Can somebody help me with second part of proof imgur.com/7qqsjP5
I don't understand how $\mu(A) = \lambda_k(A+x) \implies \lambda_k(A) = \lambda_k(A+x)$ I got the fact that $\mu$ & $\lambda_k$ agree on $\mathcal{R}^k$
@robjohn Any help??
@RamanaVenkata hang on
08:54
Greetings
@robjohn have you ever met some alternative ways of expressing $$\psi(1\pm i)$$?
@robjohn
are u still there
;)
@Chris'ssis I gave a few the other day, I think. I am looking at something for a bit; I will look for it in a bit
@robjohn let me know if you are not busy i need some help with where we ended yesterday about the distribution
@robjohn Got it.
@RamanaVenkata what is $\mathscr{R}^k$?
it doesn't seem to be $\mathbb{R}^k$
09:08
collection of borel sets
generated by rectangles
@RamanaVenkata Ah... that is kind of essential
@RamanaVenkata There is just so much undefined there. What is $\lambda_k$?
@RamanaVenkata Ah, so I can stop thinking about it? :-)
@Chris'ssis Now I can look for that :-)
sorry $\lambda_k((a_1, b_1] \times (a_2, b_2] \times ... \times (a_k, b_k] )= \prod_{i=1}^k (b_i - a_i)$
I am trying understand translational invariance under the measure $\lambda_k$ on $\mathcal{R}^k$
@Chris'ssis $-\gamma+\sum_{k=1}^\infty\frac1{k(k^2+1)}+i\sum_{k=1}^\infty\frac1{k^2+1}$
@robjohn thanks. I was looking for a closed form, maybe something involving trigonometric functions?
09:24
@robjohn just a yes or no answer ;). if as yesterday we had $P(Y_1 = s) = P(X_1 = s n)$ were $Y_1 = X_1/n$ and $X_1 \sim Bin(n,1/k)$ then $P(Y_1) \sim Bin(1,1/k^n)$
@Chris'ssis The imaginary part is $\frac{\pi\coth(\pi)-1}{2}$
@robjohn yeah, true.
@Danny That doesn't look right... The numbers from the Binomial distribution are the same, but the arguments to the distribution function are altered. There is not enough detail in describing the Binomial distribution with two parameters.
@robjohn sorry i mean $X_1 = Bin(n,1/k)$ as we agreed yesterday and $Y_1 = Bin(n,1/k^n)$
$\mathrm{Bin}(1,1/k^n)$ has only two points $0$ and $1$ with probability $\frac12$
09:30
sorry @robjohn was a mistake
@Danny I don't think I ever saw that...
but do u agree with the corrected version since $1/k^n$ is the probability when $Y_1 = 1$ so now i have the distribution for $Y_1$ given by $\binom{n}{i}(1/k^n)^i (1-1/k^n)^{n-i}$
@Danny no... the distribution for $Y_1$ is essentially the same, just scaled. This distribution has different probability values.
http://math.stackexchange.com/questions/663101/for-n-geq-2-prove-that-1-frac141-frac191-frac116
so confused towards the end :/!
argh what do I do just sub a number that's greater than 2 afterwards?
09:52
you mean like a specific number?
wait, it looks like you showed P(n) implies P(n+1). now you just need P(2). have you seen induction before?
induction is like a sequence of dominoes falling. it has two components: a base case [where you knock the first domino over], and transitions from each case to the next [this is the P(n) implies P(n+1) part]
Yes it's the basis part.. .. I just plug in P(2)
and the result should be $\frac{3}{4}$
10:33
What to do when I do not know how to proceed in the given problem? Can I ask on main without showing my any efforts if I can't get any intuition regarding problem?
@usukidoll Since you have accepted an answer, I assume you are clear now?
@robjohn, I have this theorem, "Let $X$ and $Y$ have a continuous joint distribution. Suppose that $\{(x, y) :f(x, y) > 0\}$
is a rectangular region $\mathbb R$ (possibly unbounded) with sides (if any) parallel to the
coordinate axes. Then $X$ and $Y$ are independent if and only if $f(x,y)=h_1(x)\cdot h_2(y)$ holds for
all $(x, y) ∈ \mathbb R$, where $h_1$ and $h_2$ are nonnegative functions." which I can't prove. Can I ask it on main without showing any efforts? Because I can't prove it?
yeah @robjohn thanks :D
@usukidoll good :-)
I'm new to the whole proof thing @___@
I find it a lot helpful if I read the book and come on here often rather than the lecture :/
sighh... even if I do have it right, my paper is going to be sent back all because it's written like off key piano
proof writing...more like technical writing that I'm unaware of
10:41
@usukidoll It is hard to know just where your difficulty was. This is why people are sticklers for giving what you have done. However, even then, it is not always easy to see where the difficulty is. Was yours just the process of induction?
yes
I have another problem that involves induction
@robjohn, can I ask in main withot showing my efforts?
$2n^2-2n$ if I factored that it's $(2n)(n-1)$ so with induction that would be $(2n)(n-1) * 2(n+1)(n+1-1) $?
@sush I wouldn't do that unless you like to be constantly downvoted
@Sush It is hard to know how the crowd will react. I think it would be okay, but I cannot guarantee that others might not close the question. You should try. If it is closed, work some more on it and add your work to the question.
@robjohn, @usukidoll, thank you so much.
10:45
@Sush Or if you are really worried, try a bit longer until you have something to show.
full question: Many married couples arrive one couple at a time at a restaurant. As each new couple arrives, they each shake hands exactly once with everybody who arrived before them (but nobody shakes hands with one's own spouse). Prove by induction $n \in N$, the total number of handshakes that have taken place when $n$ couples are present is $2n^2-2n$. Hint : When the (n+1)^st couple arrives, how many additional handshakes take place?]
so there are n natural numbers $n \in N$
@Sush I think that it is valid to be stuck. It sometimes just requires a small hint. Unfortunately, the homework issue has forced the "no work, no help" attitude on the community.
the only thing I could think of at this point is factoring out the $2n^2-2n$ which gives me $(2n)(n-1)$
then through induction that's $P(n) \rightarrow P(n+1)$ sub in $2n^2-2n$ ugh what
@usukidoll So check it out for $n=1$
$P(1) = 2-2 =0 $
10:48
one couple, no one to shake hands, right
$P(2) = 8-4=4$
right
$P(3) = 18-6 = 12$
if you have $n$ couples and a new couple arrives, how many handshakes occur? $2n$ for each person in the new couple.
ok for $2n$
$P(0) = 0$
$P(1) = 2$
$P(2) = 4$
$P(3) =6$
So the formula gives $P(n)=2n^2-2n$ and the additional $4n$ gives $2n^2+2n$ which is $P(n+1)=2(n+1)^2-2(n+1)$
:/ I shouldn't have factored out the $2n$ in the first place x.x
yes ^^
10:52
@robjohn Is it possible that $F$ is not left continuous here imgur.com/nrdicuc the theorem 10.2 is here imgur.com/LEs2XmE and $x_n \uparrow x$ is defined as an increasing sequence $x_n$ that converges to $x$
so if $P(n)$ is any positive number, that's a lot of handshakes
for $2(n+1)^2 -2(n+1)$
@usukidoll for a large group of people
$2n^2 -2n$
$2(n+2)^2-2(n+2)$ is even more people giving handshakes
@usukidoll that would be for $n+2$ couples
@RamanaVenkata let me look
... oh
$2n^2-2n$
$2(n-1)^2-2(n-1)$
11:01
@robjohn, thank you. sorry I am commenting late due to my net problem. I asked it here with my poor try!
@RamanaVenkata consider the Dirac delta measure.
I don't know how it is defined
@robjohn I got the definition here en.wikipedia.org/wiki/Dirac_measure But I still can't see it Can you explain it??
@sush upvoted ur question
11:31
so what happens if there is 6n....
$P(6n) = 2(6n)^2 -2(6n)$ tons of handshakes..........
@usukidoll, thank you. Can you please answer me? I will be obliged.
I don't know any of that :/
@usukidoll, no problem. But, seriously, I want the answer immediately! Today evening I have to appear the exam :|
I'm not even sure if for every new couple arrives, $2n$ handshakes for each person is the final answer :/ maybe it is
I don't know much about it sorry @sush
@usukidoll, don't you think math.se has got slow? Before 3 months I used to get the first response within 5 minutes.
11:41
I think it depends on how many people know the topic and how well you try on the problem in question
@usukidoll, ok. I upvoted your answer, so you get +10.
yay :)
I remembered when I first typed the whole thing
I was like wth only 1 vote ehhhhhhhhhhhhhhh! The whole thing was correct...then the upvotes started to apper
appear
@usukidoll,Oh, very nice. I have mostly never answered anything because i don't have math confidence. I am economics undergrad. what is your acadamic status?
math major just starting out with this logic hoo ha
ugh I am dreading it...I bet a lot of students are lost too
I mean it's their first time d***n it
what is d***n here? @usukidoll
I can't understand what do you mean by d***n
11:50
@robjohn Heya
I still could not quite work out that integral.. =(
@sush censoring the word damn
@usukidoll, ok. As I am not native english speaker, I have to read a sentence at least twice!
just edited the question in hope of answer.
Any set theorists lurking in here?
@N3buchadnezzar okay... I thought I put the whole thing in chat. Hang on.
@robjohn would the answer just be that for every handshake, there's $2n$ people?
11:55
@robjohn, can you please answer me here?
a) $u$ is the real part of $ze^z$
b) $v=xe^x\sin(y)+ye^x\cos(y)$
c) $\int_0^{2\pi}h(\theta)\,\mathrm{d}\theta=\mathrm{Re}\left(\int_0^{2\pi}e^{i\theta}e^{e^{i\theta}}\frac{\mathrm{d}e^{i\theta}}{ie^{i\theta}}\right)$
@usukidoll yes. That counts all the handshakes.
ok thanks... @robjohn
I'll just type that out .... after I wake up lol
I should be sleeping by now

« first day (1280 days earlier)      last day (4035 days later) »