New posts

Best of luck to the class of 2025 for their HSC exams. You got this! Let us know your thoughts on the HSC exams here

HSC 2015 MX2 Marathon ADVANCED (archive) (1 Viewer)

Thread starter Trebla
Start date Dec 6, 2014

Status: Not open for further replies.

Trebla

Administrator

Administrator

#1

Post any questions within the scope and level of Mathematics Extension 2 mainly targeting Q16 difficulty in the HSC.

Any questions beyond the scope of the HSC syllabus should be posted in the Extracurricular Topics forum:
http://community.boredofstudies.org/238/extracurricular-topics/

Once a question is posted, it needs to be answered before the next question is raised.

I encourage all current students in particular to participate in this marathon.

I will get the ball rolling:

Prove that

$\dfrac{3}{8}+\dfrac{4}{16}+\dfrac{5}{32}+\dfrac{6}{64}+....= 1$

Last edited: Dec 7, 2014

SilentWaters

Member

#2

Re: HSC 2015 4U Marathon - Advanced Level

Denote the summation on the LHS as $S$ so that:
$2S = \frac{3}{4} +\frac{4}{8} + \frac{5}{16} + \dots$

By subtraction:
$S = \frac{3}{4} + \left(\frac{4}{8}-\frac{3}{8}\right) + \left(\frac{5}{16}-\frac{4}{16}\right)+\dots\\=\frac{3}{4}+\frac{1}{8}\sum\limits_{k=0}^{\infty}{\frac{1}{2^n}}$
which by evaluating the limiting geometric sum equates to $\frac{3}{4} + \frac{1}{8}\cdot 2 = 1.$

Sy123

This too shall pass

#3

Re: HSC 2015 4U Marathon - Advanced Level

$\\ $Let$ \ a_0 > 0 \ $and$ \ a_{n+1} = a_n + \frac{1}{\sqrt[k]{a_n}} \\ \\ $Show that$ \ a_n \ $does not converge as$ \ n \ $increases without bound$$

-----------------------

$\\ $HINT: Consider$ \ \lim_{ n\to \infty} (a_{n+1} - a_n) \ $. Then make an interesting logical argument from there!$$

Last edited: Dec 8, 2014

Sy123

This too shall pass

#4

Re: HSC 2015 4U Marathon - Advanced Level

Sy123 said:
$\\ $Let$ \ a_0 > 0 \ $and$ \ a_{n+1} = a_n + \frac{1}{\sqrt[k]{a_n}} \\ \\ $Show that$ \ a_n \ $does not converge as$ \ n \ $increases without bound$$

-----------------------

$\\ $HINT: Consider$ \ \lim_{ n\to \infty} (a_{n+1} - a_n) \ $. Then make an interesting logical argument from there!$$

Here is my solution:

$\\ $Take the proposition that$ \ a_n \ $converges$ \\ \\ $1. If$ \ a_n \ $converges, then$ \ \lim_{n \to \infty} (a_{n+1} - a_n) = 0 \\ \\ $2. If$ \ \lim_{n \to \infty}(a_{n+1} - a_n) = 0 \ $then$ \ \left(\lim_{n \to \infty} \frac{1}{a_n} \right)^k = 0, \ $(by the recurrence relation)$ \\ \\ $3. If$ \ \left(\lim_{ n \to \infty} \frac{1}{a_n} \right)^k = 0 \ $then$ \ a_n $is divergent!$ \\ $For if it were convergent, it would converge to a value, $ \ C \ $but then$ \ 0 = \left(\lim_{n \to \infty} \frac{1}{a_n} \right)^k = C^{-k} \neq 0 \\ \\ $4. Therefore, if$ \ a_n \ $is convergent, then$ \ a_n \ $is divergent! (1,2,3)$ \\ \\ $5. Hence$ \ a_n \ $is not convergent as$ \ a_n \ $cannot both be convergent and divergent$$

F

FrankXie

Active Member

#5

Re: HSC 2015 4U Marathon - Advanced Level

Chlee1998 said:
Can someone help with Question 8 part b ii gamma?

the link to the paper is https://onedrive.live.com/view.aspx...33&app=WordPdf&authkey=!AIZ8Uo1ng7VFwpg&wdo=1

thanks to whoever can help

It is just the limiting sum formula for a GP with common ratio absolute value less than 1.

Sy123

This too shall pass

#6

Re: HSC 2015 4U Marathon - Advanced Level

$\\ $NEW QUESTION - required knowledge: Polynomials and Complex Numbers$$

$\\$Let the polynomial$ \ E(x) = 1 + \frac{x}{1!} + \frac{x^2}{2!} + \dots + \frac{x^n}{n!} \\ \\ $i) Prove that$ \ E(x) \ $has no double root$ \\ \\ $ii) Prove that if$ \ n \ $is even, then$ \ E(x) \ $has no real roots$$

seanieg89

Well-Known Member

#7

Re: HSC 2015 4U Marathon - Advanced Level

Sy123 said:
Here is my solution:

$\\ $Take the proposition that$ \ a_n \ $converges$ \\ \\ $1. If$ \ a_n \ $converges, then$ \ \lim_{n \to \infty} (a_{n+1} - a_n) = 0 \\ \\ $2. If$ \ \lim_{n \to \infty}(a_{n+1} - a_n) = 0 \ $then$ \ \left(\lim_{n \to \infty} \frac{1}{a_n} \right)^k = 0, \ $(by the recurrence relation)$ \\ \\ $3. If$ \ \left(\lim_{ n \to \infty} \frac{1}{a_n} \right)^k = 0 \ $then$ \ a_n $is divergent!$ \\ $For if it were convergent, it would converge to a value, $ \ C \ $but then$ \ 0 = \left(\lim_{n \to \infty} \frac{1}{a_n} \right)^k = C^{-k} \neq 0 \\ \\ $4. Therefore, if$ \ a_n \ $is convergent, then$ \ a_n \ $is divergent! (1,2,3)$ \\ \\ $5. Hence$ \ a_n \ $is not convergent as$ \ a_n \ $cannot both be convergent and divergent$$

The "taking the limit" step is the real key. We can pass to the limit immediately from the question statement to get:

$C=C+C^{-1/k}$

Which trivially has no real (or complex) solutions. (ie, we have the desired contradiction.)

This is all just streamlining though of course.

dan964

what

#8

Re: HSC 2015 4U Marathon - Advanced Level

** If f(nx) and g(nx) are to such that in terms of a polynomial; whereby
the sum of the roots in any amount (individually, in pairs, triples etc.) or the product of the roots is the same value - an integer
==> the coefficients of the polynomial are identical and integer values.

And also that g(x) = sqrt(1 - f(x)^2)
Show that n=5 is the first integer solution, apart from any trivial cases.

This question requires a bit of intuition.
edit: identical is wrong word maybe.

Last edited: Jan 5, 2016

Axio

=o

#9

Re: HSC 2015 4U Marathon - Advanced Level

Sy123 said:
$\\ $NEW QUESTION - required knowledge: Polynomials and Complex Numbers$$

$\\$Let the polynomial$ \ E(x) = 1 + \frac{x}{1!} + \frac{x^2}{2!} + \dots + \frac{x^n}{n!} \\ \\ $i) Prove that$ \ E(x) \ $has no double root$ \\ \\ $ii) Prove that if$ \ n \ $is even, then$ \ E(x) \ $has no real roots$$

I. E(x)-E'(x)=(x^n)/n! Which cannot equal zero. But for a double root, this must equal zero when the root is subbed in. Therefore a double root does not exist.

II. I'll try it tomorrow

F

FrankXie

Active Member

#10

Re: HSC 2015 4U Marathon - Advanced Level

Chlee1998 said:
No it's not. It's asking for the value of the coefficient not the sum of everything.
Can you post your full working?

$$ It follows from the limiting sum formula that $ F(z)=\frac{A}{1-\frac{z}{\alpha}}+\frac{B}{1-\frac{z}{\beta}}=A\sum_{k=0}^{\infty}\left(\frac{1}{\alpha}\right)^kz^k+B\sum_{k=0}^{\infty}\left( \frac {1}{\beta} \right)^kz^k=\sum_{k=0}^{\infty}\left\{A \left( \frac{1}{\alpha} \right)^k+B \left( \frac{1}{\beta} \right)^k\right\}z^k,$

$$ on the other hand, $ F(z)=\sum_{k=0}^{\infty}t_k(x)z^k$

Now equating the coefficients of like powers of these two series, we have

$t_k(x)=A \left( \frac{1}{\alpha} \right)^k+B \left( \frac{1}{\beta} \right)^k$

Last edited: Dec 9, 2014

F

FrankXie

Active Member

#11

Re: HSC 2015 4U Marathon - Advanced Level

dan964 said:
If f(nx) and g(nx) are identical polynomial of the form
A1^n + A2^(n-1) + ... + An
Ax are constants

And g(x) = sqrt(1 - f(x)^2)
Show that n=5 is the first integer solution.

I still don't understand, if g(x) = sqrt(1 + f(x)^2) and f(nx)=g(nx), isn't it that f(nx)^2=1/2 identically?

Last edited: Dec 9, 2014

Axio

=o

#12

Re: HSC 2015 4U Marathon - Advanced Level

Sy123 said:
$\\ $NEW QUESTION - required knowledge: Polynomials and Complex Numbers$$

$\\$Let the polynomial$ \ E(x) = 1 + \frac{x}{1!} + \frac{x^2}{2!} + \dots + \frac{x^n}{n!} \\ \\ $i) Prove that$ \ E(x) \ $has no double root$ \\ \\ $ii) Prove that if$ \ n \ $is even, then$ \ E(x) \ $has no real roots$$

Axio said:
I. E(x)-E'(x)=(x^n)/n! Which cannot equal zero. But for a double root, this must equal zero when the root is subbed in. Therefore a double root does not exist.

II. I'll try it tomorrow

II. E'(x)=1+x+x^2/2!...+x^(n-1)/(n-1)!
At stationary point, E'(x)=0, -1=x+x^2/2!...+x^(n-1)/(n-1)!
Therefore at stationary point E(x)=1-1+x^n/n!=x^n/n!, and as n is even, E(x) is positive at any stationary point and as the polynomial is of even degree and has a positive leading coefficient, this means that the graph is always above the x-axis and therefore has no real roots.

Last edited: Dec 9, 2014

Sy123

This too shall pass

#13

Re: HSC 2015 4U Marathon - Advanced Level

Axio said:
II. E'(x)=1+x+x^2/2!...+x^(n-1)/(n-1)!
At stationary point, E'(x)=0, -1=x+x^2/2!...+x^(n-1)/(n-1)!
Therefore at stationary point E(x)=1-1+x^n/n!=x^n/n!, and as n is even, E(x) is positive at any stationary point and as the polynomial is of even degree and has a positive leading coefficient, this means that the graph is always above the x-axis and therefore has no real roots.

Yep that's the right idea

A more rigorous way of doing so would be

"if E has no double roots, then each root is a single root, and so then there is some value 'c' So that E(c) < 0, but E(x) -> +inf as x grows large and E(x) -> +inf as x becomes negatively large, so, there must be another root, lets call these pair of roots A and B, since E(A) = E(B) = 0, then there is a point C where E'(C) = 0 And E(C) < 0, but as you have shown, E(C) > 0, so there is no root"

Sy123

This too shall pass

#14

Re: HSC 2015 4U Marathon - Advanced Level

$\\ $Let$ \ p(x) \ $be a polynomial with degree$ \ n \ $with real roots$ \ \alpha_1, \alpha_2, \dots \alpha_n \\ \\ $Prove that:$ \ p'(x) = \frac{p(x)}{x - \alpha_1} + \frac{p(x)}{x - \alpha_2} + \dots + \frac{p(x)}{x - \alpha_n}$

C

Chlee1998

Member

#15

Re: HSC 2015 4U Marathon - Advanced Level

Sy123 said:
$\\ $Let$ \ p(x) \ $be a polynomial with degree$ \ n \ $with real roots$ \ \alpha_1, \alpha_2, \dots \alpha_n \\ \\ $Prove that:$ \ p'(x) = \frac{p(x)}{x - \alpha_1} + \frac{p(x)}{x - \alpha_2} + \dots + \frac{p(x)}{x - \alpha_n}$

Rhs is (x-a2)(x-a3).....(x-an)+(x-a1)(x-a3)(x-a4)....(x-an).+...+ (x-a1)(x-a2).... (xn-1)
And its easy to see that this is p'(x) through extended product rule

Last edited: Dec 10, 2014

Sy123

This too shall pass

#16

Re: HSC 2015 4U Marathon - Advanced Level

Chlee1998 said:
Rhs is (x-a2)(x-a3).....(x-an)+(x-a1)(x-a3)(x-a4)....(x-an).+...+ (x-a1)(x-a2).... (xn-1)
And its easy to see that this is p'(x) through extended product rule

In HSC exams I would think that you'd need to justify the use of the "extended product rule", so how would you do it?

seanieg89

Well-Known Member

#17

Re: HSC 2015 4U Marathon - Advanced Level

Sy123 said:
In HSC exams I would think that you'd need to justify the use of the "extended product rule", so how would you do it?

Where p is nonzero we have

$\frac{p'}{p}=(\log |p|)'=(\sum_i \log |x-\alpha_i|)'=\sum_i \frac{1}{x-\alpha_i}.$

The result extends to all x by continuity.

Sy123

This too shall pass

#18

Re: HSC 2015 4U Marathon - Advanced Level

$\\ $A polynomial$ \ p(x) \ $has integer coefficients,$ \ p(x) \ $also has a root of the form$ \\ p + \sqrt{q} \ $where$ \ p, q \ $are rational and$ \ q \ $is not a square of a rational$ \\ \\ $Prove that$ \ p(x) \ $also has a root of the form$ \ p -\sqrt{q}$

SilentWaters

Member

#19

Re: HSC 2015 4U Marathon - Advanced Level

Sy123 said:
$\\ $A polynomial$ \ p(x) \ $has integer coefficients,$ \ p(x) \ $also has a root of the form$ \\ p + \sqrt{q} \ $where$ \ p, q \ $are rational and$ \ q \ $is not a square of a rational$ \\ \\ $Prove that$ \ p(x) \ $also has a root of the form$ \ p -\sqrt{q}$

For non-zero $p$ :

If $p(x)$ has integer coefficients, then the product of its roots must be of a rational form (the quotient of two integers, the leading coefficient and the constant). To rationalize such a product with some irrational $p+\sqrt{q}$ we must have its conjugate as a necessary root. This comes by way of the difference of two squares identity:

$(p+\sqrt{q})(p-\sqrt{q}) = p^{2}-q,$

which is rational.

For $p = 0$ :

Express $\sqrt{q}$ as a mixed surd $Q\sqrt{Q'}$ where $Q\in\mathbb{Z}\neq 0$ and $Q'$ is coprime with the square numbers. For $Q=1$ it is obvious that we need $-\sqrt{q}$ as a root for a rational sum of roots one at a time. For $Q\neq 1$ we could negate the irrational $\sqrt{q}$ with the sum of some number of smaller surds greater than 1. We are then dealing with at least 3 potential roots. Any summation of roots an odd number at a time (from 3 onwards) would produce an irrational product retaining $\sqrt{Q'}$ . The only way we can eliminate the irrational part in every case of summation would be to refrain from decomposing that root in the first place, and qualifying $-\sqrt{q}$ as a root. This is because taking cyclic sums will always either directly multiply the two simple surds to give the rational $-q$ or multiply both by the same factors (comprising other roots), allowing one to factorise out $\sqrt{q}-\sqrt{q} = 0$ and thus eliminate the irrational.

Last edited: Dec 10, 2014

seanieg89

Well-Known Member

#20

Re: HSC 2015 4U Marathon - Advanced Level

SilentWaters said:
To rationalize such a product with some irrational $p+\sqrt{q}$ we must have its conjugate as a necessary root.

You should probably provide more justification for this statement.

Status: Not open for further replies.

Share:

Facebook Twitter Reddit Pinterest Tumblr WhatsApp Email Link

Users Who Are Viewing This Thread (Users: 0, Guests: 1)

Top