« first day (1154 days earlier)      last day (4161 days later) » 
00:00 - 17:0017:00 - 00:00

00:24
I know the chance is small but anyone here is an actuary?
00:51
ho ho ho ho
santa holds some special gifts for all
manthanomen
01:14
hi Danny
hi
hey there twink
good to see your smiling face
:D lol
any good math conundrums hit you today?
01:20
mabye
I'm about to start my homework
@anon I have a conundrum.
01:37
@PedroTamaroff no entendi el chiste de Bart
me explicas en español?
Only one conundrum @Pedro?
@TedShifrin Well, at the moment... =)
I think I got it, though.
Ah ... Good.
@PedroTamaroff why do you ignore me?
@Twink: Isn't it the middle of the night there?
01:43
@TedShifrin It's about finding the determinant of the matrix $A_{ij}=1-\delta_{ij}$ over $\Bbb Z_2$.
I'm a vampire
Ted
@Twink It's about how everyone around Bart gets sucked into a black hole of ignorance.
Well, that's no surprise, @Twink :)
so you were telling me I'm ignorant?
@Twink Ugh, no. I was making a parallelism, between Bart's misbehaviour in school and your misbehaviour here.
01:45
:(
I behave well
Interesting, @Pedro. What is your approach?
only one person has answered my +50 bounty question
and it's not a satisfacotry answer
:-(
@TedShifrin Make the following row operations $R_{i}+R_{i+1}\to R_i$ for $i=1,2,\ldots,n-1$, and $R_1+R_n\to R_n$. This when $n>2$.
You get a matrix with "elevens" in the diagonal, and a lonely 1 down left.
ew row operations
Its det is easily seen to be $0$.
01:47
Elevens?
@TedShifrin Yep.
Like
11000
01100
00110
00011
10001
Then expand through the first column say.
Oh ... I won't try it your way. I would do induction or use something you might not know yet.
@TedShifrin OK?
Well, I know you know it, but I don't know your course knows it yet.
@TedShifrin OK?
01:51
@pedro are you sure that's right? Seems to me the determinant is n-1 mod 2
You're repeating yourself :)
@AnthonyCarapetis Dunno. Maybe I messed up.
do you guys like Glee?
I want to think characteristic polynomial.
@Twink Not really, no.
01:51
why not?
this song is adorable youtube.com/watch?v=-Si-GfzcCkc
It's fairly easy to get $(n-1)^n$ from the definition of the determinant since the sign of the permutation is irrelevant in Z/2
Maybe my idea isn't so bright. ponders
Ted do you like that song?
they are a gay couple
the guys singing
Sorry, @Twink: I'm thinking math now.
but music can relax you :D
01:58
@Pedro: Yes, I like my approach. I agree with @Anthony, although I got $n+1\pmod 2$.
@TedShifrin Well, let me find my mistake.
Pedro did you watch the video?
Isn't the determinant of

1 0 0 0 0
1 1 0 0 0
0 1 1 0 0
0 0 1 1 0
0 0 0 1 1

equal to 1?
@Twink Nope.
the determinant of any triangular matrix is equal to the product of its diagonal entries
@anon Well, yes, I just wanted a sanity check.
02:04
why?
Well, using my approach I am getting that if the above is A, for n=6 the det is det A^t-det A=1-1=0.
I listened to your music yesterday
@anon do you like that song?
Oh, derp.
I used up the same row twice.
OK, now it checks out.
For n=6 using good operations I get an equivalent

1 1 0 0 0 0
0 1 1 0 0 0
0 0 1 1 0 0
0 0 0 1 1 0
0 0 0 0 1 1
0 0 0 0 0 1
When n is odd, I will get a row of zeroes at the end.
Good.
So 1 if n is even, 0 if n is odd.
@TedShifrin How did you solve it?
Think of the characteristic polynomial of the matrix with $1$s everywhere!
@TedShifrin Let me try.
So $(1,\ldots,1)$ is an eigenvector with eigenvalue $n$.
02:20
Yup. And the other eigenvalues?
@TedShifrin Let me see.
@TedShifrin $(1,0,\ldots,1,0)$ is another eigenvector.
For $n$ even at least.
@TedShifrin How should I find the eigenvalues without finding the eigenvectors?
I am sad:(
You should indeed @Pedro. What's rank?
@TedShifrin dim of the image.
Which is $1$ here.
Ok, so $(X-n)^n$?
Nooooooo.
Yes @ 1.
02:26
No?
Oh, wait.
Then it would have rank $n$.
@TedShifrin Why?
If rank is $1$, what's $\dim\ker$?
@TedShifrin Oh, wait. So it has $0$ has an eigenvalue too. How boring.
So $(X-n)X^{n-1}$?
ROFL
Yes. Now finish :)
02:29
Well, put $X=1$.
Yes, in $\Bbb Z_2$.
Cool, eh?
Sure.
@TedShifrin Well, since I am not such a pro with eigenvalues and whatnot I chose my simpler path, but it is nice, yes.
You'll develop more tools with practice. I hate row ops if I don't need 'em, although I start teaching them on Friday.
@TedShifrin I found them really useful here!
I don't think I've used a row op since my linear algebra exam
02:33
Oh, derp.
They don't want the det.
They want the range.
I misread.
Now I am fucked.
They're useful for homology computations, @Anthony.
LOL.
I am a total newb to range, really.
Use columns instead of rows, if you like, @Pedro. Otherwise, you can get it from row echelon form.
02:37
Well, from the work above we know it has full rank if $n$ is even.
=D
Then you're half done. :)
I am thinking it has rank $n-1$.
Because of my computation.
Greta Garbo and Monroe, Dietrich and DiMaggio, Marlon Brando, Jimmy Dean, On the cover of a magazine
Yep.
I can get $I_{n-1}$ inside easily @Ted. Yay.
I agree @Pedro. So the image/range is a hyperplane. What's the equation for it?
02:53
@TedShifrin $(x,x,\ldots,x)$?
that's no hyperplane!
Is it Superman?
Hypersuperman.
@TedShifrin The image is $\langle (1,1,1,\ldots,1)\rangle$ is it not?
02:54
That's the perp of it, silly goose.
@TedShifrin Oh, sorry, I was thinking about your other matrix.
The one with all ones in it.
Right. It doesn't apply here :)
you chat a loooooot with each other
@TedShifrin Well, as you just said the image is the orthogonal complement of the kernel of the transpose so $\langle (1,1,1,1,\ldots,1)\rangle ^{\perp}$.
All this thinking made me hungry.
Bedtime for me. Advising appointment early morning.
Night, all.
03:00
a special good-bye to Pedro? Ted
or that's all?
@TedShifrin Nighters.
03:25
love is in the air
Now that's actually a good song @Twink
Probably before your time though
03:43
@J.W.Perry I like Glee
I like Spyro Gyra and various other sundries.
do you like Lady Gaga?
A bit after my time. I cannot really comment on Lady Gaga, sorry.
ok:)
she can sing jazz too youtube.com/watch?v=4NmSJcWowjo
Hey Twink this in not bad! She can sing.
03:53
of course she can :)
This is an old jazz standard
Ok @Twink, I heard yours, now you check out this old Hi-Lo's arrangement tenderly with parts sung in separate locations by some pretty brilliant music majors.
wow it's cool
and without any musical instrument
only their voices
:))) I am glad you like it.
04:09
now watch this video youtube.com/watch?v=2ceFnlD68Uc if you're strong enough to watch it
04:20
Your'e killin me here man. Dogs playing. Ok I watched it, so now you have to watch this cheezy assed pomeranian video. I espescially like the song. This old lady's son is a musician, and rewrote this score electronically (I like making stupid crap in Ableton so I dig his style. I also own poms).
Another mathematical conundrum would be nice about right now.
lol what a cute dog
I love pomeranian dogs
I'm gonna have one someday
They are really cool dogs.
They like to announce activity by barking though. It is in their nature, and trying to stop it is futile, possibly cruel. Great watchdogs.
04:39
you know about dogs :P
Just pomeranians by virtue of living with them for years.
oh that's cool, then I'll ask you for some advice when I have one xD
indeed :)
I'm gonna go to sleep now
good night
niters Twink
05:16
Is this complex valued integral correct?
Notice that by partial fraction decomposition we have,

$$
\int_{\gamma} \frac{1}{z^2-2z}\ dz = \frac{1}{2}\int_{\gamma} \frac{1}{z}\ dz - \frac{1}{2} \int_{\gamma} \frac{1}{z-2}\ dz.
$$

We know by the Fundamental Theorem of Calculus for Contour Integrals that since $\gamma$ is a closed path and since the principal branch of the log is analytic on some open set containing $\gamma$ then $(1/2) \int_{\gamma}(1/z)\ dz = 0$. For the other part we can see, that in a similar fashion of Problem 2 part (c), we parameterize $\gamma: [0,2\pi] \to \mathbb{C}$ as $ \gamma(\theta)= e^{i\theta} + 2$ a
 
2 hours later…
07:10
Greetings
hello
What is the partition set $G/H$ ???

is it $\lbrace x H; x\in G\rbrace $ ???
please help me
Another nice product to compute $$\lim_{n\to\infty}\prod_{k=2}^{n} \left(1-\frac{1}{4n^2 \log\left(1+\left(\displaystyle \frac{k}{n}\right)^2\right)}\right)$$
@Chris'ssis how to write $G/H$ please
 
1 hour later…
08:34
Hi all , I have a simple doubt regarding unit impulse function.
Can I ask it
09:26
Law of the unconscious statistician: Proof for the continuous random variable case. Hints please.
 
1 hour later…
10:34
hehe, amazing things happen here. I computed the limits without pen and paper and created a bunch of new ones.
@Vrouvrou I'm rather out of practice with abstract algebra. Moreover, when I attended it my professor had to get used to my own notations. Sometimes I don't follow the rules.
@robjohn are you around?
@Chris'ssis yes
@robjohn do you like my product above? I'm trying to find an elementary way to finish it beautifully.
@Chris'ssis looking now.
@robjohn ok. Hope you like it.
10:49
brb
11:00
@Chris'ssis $\frac{8}{3\pi}$?
@robjohn precisely.
@Chris'ssis I used Stirling. I don't know if there is anything cleaner.
@robjohn Interesting. I didn't think of Stirling.
@Chris'ssis How did you do it?
11:08
@robjohn I'm thinking to split that product at $\lfloor \sqrt{n}\rfloor+1$.
@robjohn then, for the $1$st product we can use elementary limits. The $2$nd product should tend to $1$.
Hmm... I used monotone convergence to
$$
\prod_{k=2}^\infty\left(1-\frac1{4k^2}\right)
$$
Turned that into a ratio of factorials, and used Stirling
@robjohn this is almost Wallis product :-)
monotonicity can be proven by Bernoulli
11:29
ho ho ho
@Danny It's not even Halloween yet.
well i still got your halloween-gift
robjohn
@Danny what was that?
here you go
see, you have been a good boy this year!
just kidding, mate. what up?
@Danny Just getting ready to work in a couple of hours. It is only 4:30 here
@JasperLoy: hey there
user87637
11:36
@robjohn You are always up very early.
what do you do
what is your job
mathematican?
@Danny I used to be, now I am a programmer and amateur mathematician.
i see. iam doing a course in programming right now
basic course
we learn scheme and python
@Danny Never learned those. I used to work in assembly and C, now I mostly use Java
are you in USA?
11:46
@Danny Yep
what do a programmer earn ? in general, i guess it depends on experience and education
but anyhow
@Danny Depends on the employer, too. Some jobs are good and some are not.
okey.
@Danny The real embarrassing thing is to be the best in your company and to be paid poorly... and when I said the best I also mean you cannot be compared to that from the 2nd position because of the high difference
are you referring to yourself? chris
;O
11:54
@Danny just felt to say that
what do u do chris
@Danny trying to be the best in anything I do.
@Chris'ssis i mean what is your occupation
@Danny freelancer (I can't tell you more)
come an
actuary,programmer ?
freelancer what is that
sorry iam from sweden i dont really get all things in english
12:02
A freelancer, freelance worker, or freelance is a person who is self-employed and is not committed to a particular employer long term. These workers are sometimes represented by a company or an agency that resells their labor and that of others to its clients with or without project management and labor contributed by its regular employees. Others are completely independent. "Independent contractor" would be the term used in a higher register of English. Fields in which freelancing is common include: music, journalism, publishing, screenwriting, filmmaking, acting, photojournalism, cosme...
photojurnalism
;)
@Chris'ssis what time is it over there
One more close vote needed here. Thanks!
@MartinSleziak done
for $f(x) = x^2$ and if A is the interval (=
damn
for $f(x) = x^2$ and if A is the interval $[0,4]$ and B is the interval $[-1,1]$ does $f^{-1}(A\cap B)$ = $f^{-1}(A)\cap f^{-1} (B)$?
12:14
It holds for any function.
For example this question is about that identity.
formally, how to show?
You can find some proofs in the answers to that question.
ok let me see
a
martin how did you find it so fast
i mean that there were such a question
i have a hard time "searching" on the forum
@Danny I looked here.
thanks martin!
12:27
Law of the unconscious statistician: Proof for the continuous random variable case. Hints please.
But Google search might work, too: For example searching for "f^{-1}" "\cap" returns a few relevant results.
And you can restrict the search to this site, if needed: "f^{-1}" "\cap" site:math.stackexchange.com
ohh
ok
 
3 hours later…
15:46
@anon
16:00
Anyone nows how to answer my question? math.stackexchange.com/questions/511207/…
for $f(x)=x^2$ and if A is the interval $[0,4]$ and B is the interval $[−1,1]$ does $f^{−1}(A∩B) = f^{−1}(A)∩f^{-1}(B)$ .NOTE: sorry for repeating but how the f*** can $f^{-1}(B)$ be defined
$y= f(x) = x^2$ but then $f^{-1} = \vert \sqrt y \vert$
:S
:(
Danny $f^{-1}$ is defined for all functions
even if it has no inverse
don't confuse the inverse of a function with $f^{-1}(A)$ for any set $A$
so you mean it is just the set
16:13
yes it's just a set
it has nothing to do with the inverse function
then what is $f^{-1}(B)$ in this context
see the definition
$f^{-1}(B)=\{x : f(x) \in B\}$
the set of all x such that $f (x) = y$
$y in/b}
which $x$'s are mapped to $B$?
0 trough 1
16:19
see the graph of $f$ to make it easier
$[0,1]$
well that's a part of $f^{-1}(B)$
$[0.1] \subset f^{-1}(B)$
$[-1,1]$
:O
yes
and for the set A $[-2,2]$
16:24
also see the graph of $f$
wich $x$'s are sent to $A$?
oh yes
that is the set
well it is that
yes
;O
twink u are my private teacher!
ok you can pay me by western union
Post a question on quasi-separated morphisms and you barely get any views...
16:35
;)
Anyone good in numerical analysis here ?
I'm trying to solve this exercise: http://math.stackexchange.com/questions/511099/what-are-the-weights-of-the-quadrate-formula-with-weight-function-x-mapsto-1-x
quality an actuary should have?
..knowing the height of a triangle..
16:51
Danny are you a girl of a guy
or*
guy
and how old are you?
29
old..
ok
you just started studying math?
or what do you study?
1,5 year, math
00:00 - 17:0017:00 - 00:00

« first day (1154 days earlier)      last day (4161 days later) »