Login

Welcome, Guest. Please login or register.

April 19, 2024, 06:20:26 pm

Author Topic: First Year University Mathematics Questions  (Read 10404 times)  Share 

0 Members and 1 Guest are viewing this topic.

goodluck

  • Adventurer
  • *
  • Posts: 18
  • Respect: 0
Re: First Year University Mathematics Questions
« Reply #30 on: February 18, 2019, 09:09:13 pm »
0
Hey,

It took me a while to find this question page but I just have this question and was wondering how you do it via vectors. I didn't really understand any of the online working out!

Prove using vector methods that the midpoints of the sides of a convex quadrilateral form a parallelogram.

RuiAce

  • ATAR Notes Lecturer
  • Honorary Moderator
  • Great Wonder of ATAR Notes
  • *******
  • Posts: 8814
  • "All models are wrong, but some are useful."
  • Respect: +2575
Re: First Year University Mathematics Questions
« Reply #31 on: February 18, 2019, 09:35:17 pm »
+5
Hey,

It took me a while to find this question page but I just have this question and was wondering how you do it via vectors. I didn't really understand any of the online working out!

Prove using vector methods that the midpoints of the sides of a convex quadrilateral form a parallelogram.

The one thing that you might not fully understand due to limitations of the HSC is what a "position vector" is yet. For the sake of these proofs we can assume that the space we're working in is some Euclidean space, and to simplify things further we can assume it's just \( \mathbb{R}^2\) or \(\mathbb{R}^3\). (You can interpret these as the usual 2-dimensional Cartesian plane, and the 3-dimensional Cartesian space. The 3D version basically also has a \(z\)-axis.)

A position vector is essentially the vector drawn from the origin to the point you're interested in. (i.e. the vector \( \overrightarrow{OP}\).) Usually, if we have a point \(P\), its position vector is denoted by \(\mathbf{p}\). You're welcome to think of it as an arrow drawn from the origin to \(P\).

However, if any point is somehow related to \(P\), its position vector is related to \(\mathbf{p}\). For example, if \(P\) was the point \( (2,3)\), and you wanted the position vector of \((4,6)\), it would actually be \(2\mathbf{p}\) because you doubled both the \(x\) and \(y\) components. If you look on a diagram, the vector from \((0,0)\) to \((4,6)\) should also look exactly twice as long as the one to \((2,3)\), and also pointing in the same direction.

Everything else, however, you kinda need to learn. Anyway here's the proof.
\[ \text{Let }ABCD\text{ be a convex quadrilateral and suppose that}\\ A,\, B,\, C\text{ and }D\text{ have position vectors}\\ \mathbf{a}\, \mathbf{b},\, \mathbf{c}\text{ and } \mathbf{d}\text{ respectively.} \]
\[ \text{Then the midpoints have position vectors}\\ M_{AB} \text{ being }\frac12 (\mathbf{a}+\mathbf{b})\\ M_{BC}\text{ being }\frac12 (\mathbf{b}+\mathbf{c})\\ M_{CD}\text{ being }\frac{1}{2}(\mathbf{c}+\mathbf{d})\\ M_{CD}\text{ being }\frac12(\mathbf{d}+\mathbf{a})\]
This is not a coincidence. This is actually just the midpoint formula you learn in 2U in disguise.
\[ \text{Now, recall that given position vectors}\\ \text{the vector }\overrightarrow{AB}\text{ is the same as the vector }\mathbf{b}-\mathbf{a}.\]
This is a consequence of vector addition. Using tip-to-tail vector addition, we can rewrite \(\overrightarrow{AB} = \overrightarrow{AO} + \overrightarrow{OB} =- \overrightarrow{OA} + \overrightarrow{OB} = -\mathbf{a} + \mathbf{b}\). Note that flipping a vector is essentially the same as taking negatives of vectors.
\[ \text{So in our case, we have}\\ \begin{align*} \overrightarrow{M_{AB}M_{BC}} &= \frac12 (\mathbf{b}+\mathbf{c}) - \frac12 (\mathbf{a}+\mathbf{b})\\ &= \frac12 (\mathbf{c} - \mathbf{a})\end{align*}\]
\[ \text{Similarly,}\\ \begin{align*}\overrightarrow{M_{CD}M_{DA}}&= \frac12 (\mathbf{d}+\mathbf{a}) - \frac12 (\mathbf{c} +\mathbf{d}) \\ &= \frac12 (\mathbf{a}-\mathbf{c})\end{align*}\]
\[ \text{But consequently }\overrightarrow{M_{AB}M_{BC}} = -\overrightarrow{M_{CD}M_{DA}}. \]
At this point we are essentially done. Note that the line segments \( M_{AB}M_{BC}\) and \(M_{CD}M_{DA}\) are opposite sides of the quadrilateral, that we're hoping to prove is a parallelogram. But if the vectors are negatives of each other, that means that they must have the same magnitude, and only point in opposite directions.

The vectors pointing in opposite directions doesn't matter though. They're still parallel as well.

So consequently, we essentially have a pair of equal and opposite sides. This is one of the many tests to prove that a quadrilateral is a parallelogram.

Note: This is definitely not the only approach. It's just my preferred approach.
« Last Edit: February 19, 2019, 08:40:36 pm by RuiAce »

justwannawish

  • Forum Obsessive
  • ***
  • Posts: 264
  • Respect: +41
Re: First Year University Mathematics Questions
« Reply #32 on: February 24, 2019, 07:51:13 pm »
0
Hey, I just have two questions, one of which had been posted on the forum previously. I'm a bit confused about the difference between the two solutions of 32b and 32c if done algebraically. I read and understood the solution for 32b but I'm not sure what the difference the new domain would create

for c) x will have to be smaller than y, right, so if we compute it out, wouldn't it still give up the same algebraic solution as b) did? or would the signs be flipped as we are considering it from the negative side?
I also tried to graphically solve it and have attached it. Could you see if I'm on the right track?

My second question is also attached. I'm just a bit unsure of how it works though I get why theta=0 is a solution ofc.


RuiAce

  • ATAR Notes Lecturer
  • Honorary Moderator
  • Great Wonder of ATAR Notes
  • *******
  • Posts: 8814
  • "All models are wrong, but some are useful."
  • Respect: +2575
Re: First Year University Mathematics Questions
« Reply #33 on: February 24, 2019, 08:09:25 pm »
+3
Hey, I just have two questions, one of which had been posted on the forum previously. I'm a bit confused about the difference between the two solutions of 32b and 32c if done algebraically. I read and understood the solution for 32b but I'm not sure what the difference the new domain would create

for c) x will have to be smaller than y, right, so if we compute it out, wouldn't it still give up the same algebraic solution as b) did? or would the signs be flipped as we are considering it from the negative side?
I also tried to graphically solve it and have attached it. Could you see if I'm on the right track?

My second question is also attached. I'm just a bit unsure of how it works though I get why theta=0 is a solution ofc.


Firstly note that your rearranged condition is off. You're right about that \(y^2 < x^2 + 1\), but this quadratic inequality solves to give \( \boxed{-\sqrt{x^2-1} < y < \sqrt{x^2+1}} \). On Desmos, you can just type the original condition \(x^2-y^2 < 1\) and the correct plot will still show.

Anyway. The order quantifiers mean everything here. Whilst it turns out both statements are true, the positioning of the quantifiers mean that they mean different things.

The first says that for all real numbers \(x\) there exists some \(y\) where \( y > x\), such that \(y^2-x^2 < 1\).
This means that \(x\) has to be any arbitrary real number. And if the statement is true, we should be able to pick some \(y\) (very likely in terms of \(x\)) that will satisfy the condition \(y^2 - x^2 < 1\).

The second says that there exists a real number \(y\) (not necessarily unique), such that for all \(x\) where \(x < y\), such that \(y^2-x^2 <1\).
This means that this time, we need to choose some real number \(y\) (and it can't be in terms of anything). That \(y\), then has to guarantee us that regardless of what \(x\) is, the condition \(y^2-x^2<1\) is satisfied.

So for the first one, you have to start by assuming \(x\) is whatever real number you want it to be. Then, you need to choose some \(y\in (x,\infty)\) that will work. One example that will work is \( y = \sqrt{x^2 + \frac12} \) - not sure what the answers they give is but that's an example of one of them.
(On Desmos, you can try plotting the required condition \(y>x\), the condition that we want i.e. \( y ^2<x^2+1\), and the curve \(y = \sqrt{x^2+\frac12}\). You'll notice that the curve lies completely between where the two regions intersect, thus affirming this answer is valid.)

Whereas for the second one, you need to commence by choosing some \(y\) that is guaranteed to always ensure that if \(x < y\), then \(y^2-x^2 < 1\). \(y=0\) should be an intuitively obvious answer here, because then \(y^2 - x^2 = -x^2\), which is always less than 1 regardless of what \(x\) is. (The condition that \(x\in (-\infty,y)\) coincidentally happens to not be that important, because it actually holds for all \(x\in \mathbb{R}\) anyway.)
(This time, on Desmos, to affirm this result we have to plot the line \(y=0\). What we want to check is that no matter what \(x\) is, the condition \(y^2-x^2 < 1\) is satisfied. Converting this graphically, we want to therefore check that the line \(y=0\), is always inside the region \(y^2-x^2 < 1\) (and doesn't, say, suddenly pop out of it at some value of \(x\). You can check that this will indeed be the case.)

Note that the condition is always \(x < y\) - this stays the same more or less as a consequence of how the questions were written. But the approaches differ because of how the quantifiers are provided.
______________________________________________

With the other question, you can just note that \(y=\sin \theta\) is contained entirely in the 1st and 3rd quadrants in that domain, whilst \(y = m\theta\) is contained entirely in the 2nd and 4th quadrants under the condition that \(m < 0\). Also note that whilst \(\sin \theta = 0\) when \(\theta = -\pi, 0, \pi\), on the other hand \(m\theta = 0\) only when \(\theta = 0\).
« Last Edit: February 27, 2019, 10:20:43 am by RuiAce »

justwannawish

  • Forum Obsessive
  • ***
  • Posts: 264
  • Respect: +41
Re: First Year University Mathematics Questions
« Reply #34 on: February 28, 2019, 05:56:20 pm »
0
It took me a while but I actually fully understand that so thank you for your extremely in-depth explanation :)

I was trying to figure out how to do this question, but I'm confused on what we do regarding the exponential/logarithm form.

Express the rational function (27x^3+1)/ (x^(3/2)) in terms of a hyperbolic function, logarithms and other functions as needed, for x>0.
« Last Edit: March 01, 2019, 06:38:40 pm by justwannawish »

RuiAce

  • ATAR Notes Lecturer
  • Honorary Moderator
  • Great Wonder of ATAR Notes
  • *******
  • Posts: 8814
  • "All models are wrong, but some are useful."
  • Respect: +2575
Re: First Year University Mathematics Questions
« Reply #35 on: March 01, 2019, 01:34:01 pm »
+2
It took me a while but I actually fully understand that so thank you for your extremely in-depth explanation :)

I was trying to figure out how to do this question on the matlab quiz, but I'm confused on what we do regarding the exponential/logarithm form.

Express the rational function (27x^3+1)/x^3/2 in terms of a hyperbolic function, logarithms and other functions as needed, for x>0.

(P.S. This is a generic thread and not just for UNSW students. Not everyone will know about the MapleTA quizzes.)

Assuming that there's no typo. Also assuming that the 3/2 is in the bottom power - please make this a bit clearer in the future.
\[ \text{Let }\boxed{e^t = \sqrt{27}x^{3/2}}\implies \boxed{t = \ln \left(\sqrt{27}x^{3/2} \right)}.\text{ Then,}\\ \begin{align*}\frac{27x^3+1}{x^{3/2}} &= \tag{clever factorising}\frac{27^{1/2} x^{3/2} \left(27^{1/2} x^{3/2} + 27^{-1/2} x^{-3/2} \right)}{x^{3/2}}\\ &= \sqrt{27} (e^t + e^{-t})\\ &= 2\sqrt{27} \cosh t\\ &= 2\sqrt{27} \cosh \left( \ln \left(\sqrt{27}x^{3/2}\right) \right) \end{align*} \]
Optionally, we may use log laws to re-express what's on the inside as \( \frac32 \ln (3x) \).

Note that the factorisation was the critical step. In general, because \( \cosh t = \frac{e^t+e^{-t}}{2}\), we need to have some \( f(x) + \frac1{f(x)}\) pattern appearing, before we can sub \(e^t = f(x)\). Same goes for \(\sinh t\) except we have a minus instead.
« Last Edit: March 01, 2019, 01:54:51 pm by RuiAce »

justwannawish

  • Forum Obsessive
  • ***
  • Posts: 264
  • Respect: +41
Re: First Year University Mathematics Questions
« Reply #36 on: March 01, 2019, 06:45:09 pm »
0
(P.S. This is a generic thread and not just for UNSW students. Not everyone will know about the MapleTA quizzes.)

Assuming that there's no typo. Also assuming that the 3/2 is in the bottom power - please make this a bit clearer in the future.
\[ \text{Let }\boxed{e^t = \sqrt{27}x^{3/2}}\implies \boxed{t = \ln \left(\sqrt{27}x^{3/2} \right)}.\text{ Then,}\\ \begin{align*}\frac{27x^3+1}{x^{3/2}} &= \tag{clever factorising}\frac{27^{1/2} x^{3/2} \left(27^{1/2} x^{3/2} + 27^{-1/2} x^{-3/2} \right)}{x^{3/2}}\\ &= \sqrt{27} (e^t + e^{-t})\\ &= 2\sqrt{27} \cosh t\\ &= 2\sqrt{27} \cosh \left( \ln \left(\sqrt{27}x^{3/2}\right) \right) \end{align*} \]
Optionally, we may use log laws to re-express what's on the inside as \( \frac32 \ln (3x) \).

Note that the factorisation was the critical step. In general, because \( \cosh t = \frac{e^t+e^{-t}}{2}\), we need to have some \( f(x) + \frac1{f(x)}\) pattern appearing, before we can sub \(e^t = f(x)\). Same goes for \(\sinh t\) except we have a minus instead.

Sorry about that! I didn't mean to restrict it to UNSW and I apologise if anyone thought that.  I've edited my above post to make it clearer (I hope I'm not bothering you or anyone who checks this thread by sending too many questions, it's just a bit tough atm, but I'll try my best to keep up with the course work!)