MAT Prep Thread 2018

Our half term is the week before, but in previous years when it has been in half term people have just had to go into their school and sit the exam in half term.

Is anyone else potentially going to be sitting the TMUA the same day? I'm looking at Durham and they encourage us to take it...but it's another 2 exams and they're all scheduled for the same day which would be a very long day

I thought Durham took the MAT into account if you were applying to Oxford?

Hmm for some reason I thought they didn't Will check!
Although I'm also looking at Lancaster and they quite like the TMUA so I'd probably end up sitting it anyway

Posted from TSR Mobile

I’m sure they do take it into account because I’ve looked into Durham but I imagine if you were to take both exams, they’d use the one with the better result or will the TMUA take precedence over the MAT

I can't shake off the idea that Oxford is easier than Cam for maths because the MAT is easier than STEP. I know STEP is harder because a) it is at the end of year 13 and b) it is in order to siphon out the last few applicants rather than eliminate the bottom half.

Still though, it's really annoying to have to make a choice between the two :/

I can't shake off the idea that Oxford is easier than Cam for maths because the MAT is easier than STEP. I know STEP is harder because a) it is at the end of year 13 and b) it is in order to siphon out the last few applicants rather than eliminate the bottom half.

Still though, it's really annoying to have to make a choice between the two :/

I would base your decision on the course content and which town you prefer. At Oxford, actually getting an interview is a hurdle (at Cambridge it's almost guaranteed unless you don't have good predicteds). Also people probably prepare far more for STEP because they have the motivation of the offer and then 5 months to work on it.

I have heard people at my school say they're going for Oxford just because it's easier - I don't necessarily agree with this. This does seem to be a recurring thing though - we always get a few into Oxford maths or CS but only one every few years into Cambridge because very few people apply there.

No worries

There shouldn't be much prerequisite knowledge for the interviews beyond the syllabus, I think they are to test application of knowledge rather than breadth of knowledge. The knowledge of graph theory required for my Lady Margaret Hall interview I believe may have only been what nodes, edges, paths and connected graphs are.



I've attached the pdf to this post (the problem sheet for Worcester College CS is attached separately as two images, I hope they're of high enough resolution) It was only the questions I had at interviews, but I'll see if other people I know have questions I can use and upload here Also, in the interviews, there was a discussion, and the questions weren't always just asked, it was often through a discussion to get the required answer. The interviewers try and help you when you are stuck and give you little hints as where your solution should be heading.



It may help to see these types of concepts, but ultimately there probably wouldn't be much knowledge from these modules required, and I assume the interviewers would be testing your application of knowledge more than the breadth

Hope this is useful

So did you get an offer and from which college (could you guess depending on your performance in the interviews)? Did you know your MAT score after the application process?

I have a really weird doubt. Is it possible to express the cube of any natural number greater than 4 , as the sum of (n-2) distinct squares?

Could you please provide the context.
Is that your conjecture you want us to prove/falsify?

I have a really weird doubt. Is it possible to express the cube of any natural number greater than 4 , as the sum of (n-2) distinct squares?

Yes. I tried it out, and it works for all numbers till 27. I will upload a photo of the result.

Just checking what you mean by n-2 there. Are you asking whether the cube of a number n can be expressed as the sum of exactly n-2 distinct squares?

Are you aware of the four squares theorem? Haven't thought about this too hard, but doesn't your conjecture follow from this by subtracting off a bunch of squares (n - 6 of them) from the n^3 and then expressing the result as four squares?

Still figuring out for 27

Figured for 27.

I am using a sort of 'recursive' idea , i.e. , I compute the value of the required squares for n^3 , find (n+1)^3-n^3 , and then suitably partition the difference into squares.

Can anyone help me with MAT 2017 question G. How do you realize that the line pivots around (1,1)