The Student Room Group

MAT Prep Thread 2019

Scroll to see replies

Original post by kasd
Is it necessary to explain how you get to the solutions of 2,3,4,5 in order to get full marks on them? or is enough having them right even though the process is messy?


Yep best to explain the longer questions so that the examiner can follow your thought process. If you're doing Oxford, then you might wanna show your working in the MCQ too because apparently they take a look at that as well! But it's just the correct answer in the MCQ that get you the marks (to clarify).
Original post by rebellionium
am I the only one who find some questions confusing? like 2011 q5(ii), I don't understand what is devise a concise way of representing the possible jorneys. I already use U and R to represent up and right moves in (i), so do the solution. I just wrote the length of a solution is n-1. How to avoid losing marks this way?

I had the same problem on that question, I had already used U and R and didn't really understand what notation they wanted.
I sometimes misunderstand the question and lose a lot of marks. It doesn't happen every paper but if it happens in the real exam I will definitely be in big trouble coz my score is marginal :frown:
Original post by rebellionium
am I the only one who find some questions confusing? like 2011 q5(ii), I don't understand what is devise a concise way of representing the possible jorneys. I already use U and R to represent up and right moves in (i), so do the solution. I just wrote the length of a solution is n-1. How to avoid losing marks this way?


For that question, I suppose its just a matter of noticing that the path lengths are all 3 and at each square you can always take two direction. So the number of solutions is 2^(n-1).
A few hints about doing it, you could have solved the n=2 problem, n=3 problem and worked up to the n=4 and you'd notice the 2, 4, 8 sequence and it should be easy to guess that its 2^(n-1). If thats the case, you could represent it as a symmetric tree where each leaf splits into two on the next level. A proof by induction would probably be overkill, but that's essentially the insight.
So
* spot magic numbers 8 = 2^3 and think about why they arise
* simplify the problem to gain insight
* if you're writing down solutions as a string, make sure you do it methodically so you can spot patterns. Each penultimate square (solution for n-1) can branch in two ways (solution for n).

For ii) you could simply say
[U/R, U/R, .... , U/R]
of length n-1 where each option is either Up or Right
(edited 4 years ago)
@mqb2766 thx for the detailed explanation but I have spotted 2^(n-1) and had written that on (iii), but I don't understand what (ii) is asking so I have only put length = n-1 there.
Original post by rebellionium
@mqb2766 thx for the detailed explanation but I have spotted 2^(n-1) and had written that on (iii), but I don't understand what (ii) is asking so I have only put length = n-1 there.


I realised at the end of the post I'd not quite answered the question, so I edited it as

For ii) you could simply say
[U/R, U/R, .... , U/R]
of length n-1 where each option is either Up or Right
Original post by rebellionium
am I the only one who find some questions confusing? like 2011 q5(ii), I don't understand what is devise a concise way of representing the possible jorneys. I already use U and R to represent up and right moves in (i), so do the solution. I just wrote the length of a solution is n-1. How to avoid losing marks this way?

With that question, there are only two possible moves from each square, U or R. As each solution must have n-1 movements, there are a total 2^(n-1) different combinations of a sequence of U and R that is n-1 long
are vectors and matrices included in the syllabus? never saw questions related to that, but I rmb one mc question can be solved by vectors, don't rmb which year tho
Original post by rebellionium
are vectors and matrices included in the syllabus? never saw questions related to that, but I rmb one mc question can be solved by vectors, don't rmb which year tho

they're not included though I think GCSE vectors should be assumed and I think I remember which question you're on about
I've been redoing questions from the past papers that I've done where I got lower marks. I've been completing them and getting quite high marks (>12/15) in 20 minutes or so, so perhaps there is hope after all...
Original post by GreenCub
How long do you all spend on the multiple choice? I find it's usually worth taking at least 45-50 minutes to try to make sure I get as many right as possible. The multiple choice is worth 40 marks, and in some past papers I've dropped 8-12 marks from being careless or misreading the multiple choice questions.


Exactly the same.
How many points do you guys get for the MCQs and long questions? I've never gotten more than four MCQs right under timed conditions so the bulk of my marks have been from the long ones, but those have been getting too hard and I can't rely on them anymore. Was only spending around 30 min on the MCQs tho, which seems like a bad idea now that I think about it.
Hey, have any of you seen 1(I) from 2016? I checked through the solution and was wondering if there was an easier (and perhaps algebraic) way to solve it. I've typed the question below too:
---
Let a and b be positive real numbers. If x^2 + y^2 ≤ 1 then the largest that ax+by can equal is:

(a) 1/a + 1/b
(b) max(a,b)
(c) sqrt(a^2+b^2)
(d) a+b
(e) a^2 + ab + b^2
---

Thanks!
Original post by asliceofpi
Hey, have any of you seen 1(I) from 2016? I checked through the solution and was wondering if there was an easier (and perhaps algebraic) way to solve it. I've typed the question below too:
---
Let a and b be positive real numbers. If x^2 + y^2 ≤ 1 then the largest that ax+by can equal is:

(a) 1/a + 1/b
(b) max(a,b)
(c) sqrt(a^2+b^2)
(d) a+b
(e) a^2 + ab + b^2
---

Thanks!

See #170
Original post by living_sandwich
How many points do you guys get for the MCQs and long questions? I've never gotten more than four MCQs right under timed conditions so the bulk of my marks have been from the long ones, but those have been getting too hard and I can't rely on them anymore. Was only spending around 30 min on the MCQs tho, which seems like a bad idea now that I think about it.

I wouldn't recommend only spending 30 minutes on the multiple choice, unless you're one of those people who thinks ridiculously quickly and manages not to make careless mistakes. Try spending more like 45 minutes (or a little more). I usually get between 7/10 and 9/10, but often under timed conditions I make careless mistakes that lowers my score.
Original post by rebellionium
are vectors and matrices included in the syllabus? never saw questions related to that, but I rmb one mc question can be solved by vectors, don't rmb which year tho

There's certainly no matrices. The MAT syllabus mentions vectors, but presumably these are only GCSE/AS level standard.

I used a matrix transformation to solve part (vi) of the MAT 2012 Q5 to find a vector that had been rotated anticlockwise by 90° (to take into account how the robot turned left before performing the same sequence of moves again) and adding together two vectors to get the final position.
Reply 196
Avatar for 1l2e
1l2e
3
Are people revising binomial hypothesis testing as part of binomial probabilities in the new syllabus?
Original post by 1l2e
Are people revising binomial hypothesis testing as part of binomial probabilities in the new syllabus?

No
Original post by asliceofpi
Hey, have any of you seen 1(I) from 2016? I checked through the solution and was wondering if there was an easier (and perhaps algebraic) way to solve it. I've typed the question below too:
---
Let a and b be positive real numbers. If x^2 + y^2 ≤ 1 then the largest that ax+by can equal is:

(a) 1/a + 1/b
(b) max(a,b)
(c) sqrt(a^2+b^2)
(d) a+b
(e) a^2 + ab + b^2
---

Thanks!


So first off, you know that x^×+y^2=1 for ax+by to be maximum (kind of obvious). Then y = sqrt(1-x^2). Sub this to get ax+bsqrt(1-x^2) and differentiate and set dy/dx=0 to find x and y that maximise it.
is the trapezium method to estimate integrals included? seems it's quite common in older papers, but can't find it in recent years

Quick Reply

Related discussions

Latest

Trending

Trending

Articles for you