Login

Welcome, Guest. Please login or register.

April 23, 2024, 09:32:39 pm

Author Topic: Beyond first-year and random math/stats questions/discussion  (Read 9200 times)  Share 

0 Members and 1 Guest are viewing this topic.

RuiAce

  • ATAR Notes Lecturer
  • Honorary Moderator
  • Great Wonder of ATAR Notes
  • *******
  • Posts: 8814
  • "All models are wrong, but some are useful."
  • Respect: +2575
Beyond first-year and random math/stats questions/discussion
« on: December 24, 2017, 10:46:33 am »
+7
Post anything beyond first year stuff here if you need help/wish to discuss/etc. :)
« Last Edit: December 26, 2017, 03:00:48 pm by RuiAce »

MLov

  • Adventurer
  • *
  • Posts: 14
  • Respect: +16
Re: Beyond first-year and random math questions/discussion
« Reply #1 on: December 24, 2017, 11:39:37 am »
+2
If p is a prime number and k is a positive integer that is greater than one. Prove cannot be expressed as a power of 2.

Can someone help me to prove this :O

RuiAce

  • ATAR Notes Lecturer
  • Honorary Moderator
  • Great Wonder of ATAR Notes
  • *******
  • Posts: 8814
  • "All models are wrong, but some are useful."
  • Respect: +2575
Re: Beyond first-year and random math questions/discussion
« Reply #2 on: December 24, 2017, 10:16:12 pm »
+9
If p is a prime number and k is a positive integer that is greater than one. Prove cannot be expressed as a power of 2.

Can someone help me to prove this :O


noting, of course, \( k > 1\).
___________________________________________


The proof of this factorisation can be done by geometric series, first with common ratio \( p\), then with common ratio \(p^2\).

This is just the basic fact that if \(x = ab\) then \(a \) and \(b \) must be factors of \(x\). Here, \(x = 2^m\) for some integer \(m\).
___________________________________________

Now for the handwavy bit.

Note: The restriction on \(N\) is enforced by the fact that \(p \ge 3\), i.e. \(p+1 \ge 4\).

where all of the \(c_i \) are assumed to satisfy \( 0 \le c_i < 2^N \)

___________________________________________

From here, a sketch will be given as the formal write-up can be messy.

In the first case, suppose that \(2^N\) is not a factor of \( \ell + 1\). Instead, suppose that \( 2^M\) is the maximum power on 2 that is a factor of \( \ell + 1\), where \(M < N\). Upon factoring out \(2^M\), equation \( (*) \) will be \(2^M\) times something.
That "something" will be the sum of a bunch of things divisible by \( 2^{N-M} \), plus something that is not (because in fact, isn't divisible by ANY power of 2). When you add something that is a power of 2 to something that is not even divisible by 2, you will get something that is not a power of 2.

Hence, the long winded expression cannot be a power of 2 either, which is the contradiction.

In the second case, where \(2^N\) is a factor of \(\ell + 1\), we rinse and repeat but shift the focus from \( \ell + 1\) to \(c_N\). (Basically, the problematic term changes.)

MLov

  • Adventurer
  • *
  • Posts: 14
  • Respect: +16
Re: Beyond first-year and random math questions/discussion
« Reply #3 on: December 25, 2017, 10:34:03 am »
+5
Christmas special
Evaluate the integral:

when
« Last Edit: December 25, 2017, 10:48:19 am by MLov »

RuiAce

  • ATAR Notes Lecturer
  • Honorary Moderator
  • Great Wonder of ATAR Notes
  • *******
  • Posts: 8814
  • "All models are wrong, but some are useful."
  • Respect: +2575
Re: Beyond first-year and random math questions/discussion
« Reply #4 on: December 25, 2017, 10:43:58 am »
+5
Christmas special
Evaluate the integral:

when

You're evil.

MLov

  • Adventurer
  • *
  • Posts: 14
  • Respect: +16
Re: Beyond first-year and random math questions/discussion
« Reply #5 on: December 25, 2017, 10:47:06 am »
+1
You're evil.

“The good Christian should beware of mathematicians. The danger already exists that mathematicians have made a covenant with the devil to darken the spirit and confine man in the bonds of Hell.”

― Augustine of Hippo

 ;)

RuiAce

  • ATAR Notes Lecturer
  • Honorary Moderator
  • Great Wonder of ATAR Notes
  • *******
  • Posts: 8814
  • "All models are wrong, but some are useful."
  • Respect: +2575
Re: Beyond first-year and random math questions/discussion
« Reply #6 on: December 25, 2017, 11:22:05 am »
+4
I mean, I'm not bothered to figure it out right now so I'm just gonna dump the mathstackexchange proof

M909

  • Forum Regular
  • **
  • Posts: 88
  • Respect: +48
Re: Beyond first-year and random math/stats questions/discussion
« Reply #7 on: April 01, 2018, 02:15:25 pm »
0
So, not totally sure if I’m allowed to bump this thread, or if there was another offical thread for this, but I’m kind of stuck with something that’s haulted my progress with probability... (Sorry if I got the wrong thread)

For ∑n*(a/b)^n from n=1 to infinity, for 0<a<b, I seem to be getting an answer of ab/(b-a)^2 through general experimenting with my calculator. Is this actually a known result, and if so does it have a name so I can look up the proof? A tutorial solution uses this result, however I don’t believe we were ever introduced to this in first year maths, and no one else in my tutorial seemed to know either... Thanks :)
VCE, 2015-2016
BCom (Econ) @ UniMelb, 2017-2019
MCEng (Elec) @ UniMelb, 2020-?

RuiAce

  • ATAR Notes Lecturer
  • Honorary Moderator
  • Great Wonder of ATAR Notes
  • *******
  • Posts: 8814
  • "All models are wrong, but some are useful."
  • Respect: +2575
Re: Beyond first-year and random math/stats questions/discussion
« Reply #8 on: April 01, 2018, 02:26:15 pm »
+4
So, not totally sure if I’m allowed to bump this thread, or if there was another offical thread for this, but I’m kind of stuck with something that’s haulted my progress with probability... (Sorry if I got the wrong thread)

For ∑n*(a/b)^n from n=1 to infinity, for 0<a<b, I seem to be getting an answer of ab/(b-a)^2 through general experimenting with my calculator. Is this actually a known result, and if so does it have a name so I can look up the proof? A tutorial solution uses this result, however I don’t believe we were ever introduced to this in first year maths, and no one else in my tutorial seemed to know either... Thanks :)

\begin{align*}\sum_{n=1}^\infty n x^n &= x \sum_{n=1}^\infty n x^{n-1}\\ &= x \sum_{n-1}^\infty \frac{d}{dx} x^n\\ &= x \frac{d}{dx} \sum_{n=1}^\infty x^n\\&= x \frac{d}{dx} \sum_{n=0}^\infty x^n \\&= x \times \frac{d}{dx} \left(\frac{1}{1-x}\right)\\ &= \frac{x}{(1-x)^2}\end{align*}
« Last Edit: April 01, 2018, 02:40:18 pm by RuiAce »

M909

  • Forum Regular
  • **
  • Posts: 88
  • Respect: +48
Re: Beyond first-year and random math/stats questions/discussion
« Reply #9 on: April 01, 2018, 04:28:41 pm »
0
Thanks a heap Rui, appreciate it :)
VCE, 2015-2016
BCom (Econ) @ UniMelb, 2017-2019
MCEng (Elec) @ UniMelb, 2020-?

swico

  • Adventurer
  • *
  • Posts: 15
  • Respect: 0
Re: Beyond first-year and random math/stats questions/discussion
« Reply #10 on: April 06, 2018, 03:32:36 pm »
0
Suppose that we have an infinite metric space, how can I show that there exists an open set where both it and its complement are infinite?

My idea was to consider the isolated points and then split it into cases where the set of isolated points are infinite or not, is this the right idea?

TrueTears

  • TT
  • Honorary Moderator
  • Great Wonder of ATAR Notes
  • *******
  • Posts: 16363
  • Respect: +667
Re: Beyond first-year and random math/stats questions/discussion
« Reply #11 on: April 06, 2018, 04:24:30 pm »
+1
You are on the right track, the case where the set of isolated points are finite is a bit tricky, but the infinite case should be quite straight forward.
PhD @ MIT (Economics).

Interested in asset pricing, econometrics, and social choice theory.

RuiAce

  • ATAR Notes Lecturer
  • Honorary Moderator
  • Great Wonder of ATAR Notes
  • *******
  • Posts: 8814
  • "All models are wrong, but some are useful."
  • Respect: +2575
Re: Beyond first-year and random math/stats questions/discussion
« Reply #12 on: April 06, 2018, 04:35:05 pm »
+1
Suppose that we have an infinite metric space, how can I show that there exists an open set where both it and its complement are infinite?

My idea was to consider the isolated points and then split it into cases where the set of isolated points are infinite or not, is this the right idea?
The proof is pretty much on stack; here I essentially give a sketch from what I infer. We basically split the cases of when our metric is discrete, or when at least one point is a limit point.

If a point is a limit point of \(X\), say \( \textbf{x} \), then for every \( \epsilon >0 \) there exists a sequence \( \{\textbf{x}_n \}_{n=1}^\infty \) with the property \( \textbf{x}_n \in B(\textbf{x},\epsilon) \quad \forall n\) and \( \lim_{n\to \infty} \textbf{x}_n \).
So pretty much just pick any ball around \( \textbf{x}\) and you're done.

Whereas if every point is isolated, then every singleton is an open set. As is its complement.

Note - don’t claim to be an expert; I only started analysis this year.
« Last Edit: April 06, 2018, 04:37:23 pm by RuiAce »

swico

  • Adventurer
  • *
  • Posts: 15
  • Respect: 0
Re: Beyond first-year and random math/stats questions/discussion
« Reply #13 on: April 06, 2018, 04:43:42 pm »
0
You are on the right track, the case where the set of isolated points are finite is a bit tricky, but the infinite case should be quite straight forward.
Thanks, I am finding the construction of the infinite case to be a bit weird, how did you construct it?

The proof is pretty much on stack; here I essentially give a sketch from what I infer. We basically split the cases of when our metric is discrete, or when at least one point is a limit point.

If a point is a limit point of \(X\), say \( \textbf{x} \), then for every \( \epsilon >0 \) there exists a sequence \( \{\textbf{x}_n \}_{n=1}^\infty \) with the property \( \textbf{x}_n \in B(\textbf{x},\epsilon) \quad \forall n\) and \( \lim_{n\to \infty} \textbf{x}_n \).
So pretty much just pick any ball around \( \textbf{x}\) and you're done.

Whereas if every point is isolated, then every singleton is an open set. As is its complement.

Note - don’t claim to be an expert; I only started analysis this year.
Thanks, is there a way without considering what metric we are using? Ie., can we prove it purely considering the infinite-ness/finite-ness of sets? It seems all the solutions on stackexchange are based off the discrete metric.

RuiAce

  • ATAR Notes Lecturer
  • Honorary Moderator
  • Great Wonder of ATAR Notes
  • *******
  • Posts: 8814
  • "All models are wrong, but some are useful."
  • Respect: +2575
Re: Beyond first-year and random math/stats questions/discussion
« Reply #14 on: April 06, 2018, 04:48:27 pm »
0
Thanks, I am finding the construction of the infinite case to be a bit weird, how did you construct it?
Thanks, is there a way without considering what metric we are using? Ie., can we prove it purely considering the infinite-ness/finite-ness of sets? It seems all the solutions on stackexchange are based off the discrete metric.
I didn't really assume what the metric was itself. The concept of an epsilon-ball holds for all metrics and metric spaces, not just the specific cases of \( \mathbb{R}^n\) and etc.