Hi,

Hindsight is a wonderful thing. Could any current 4th year MMath students, MMath graduates, or (current or previous) MSc students please give some ideas of things they wish they knew before they started their projects/dissertations, and any tips for those who plan on embarking down the same route?

I start my fourth year in October, and so far all I've heard is "Choose the supervisor, not the project.", "Learn how to use LaTeX (and Beamer)." and "Do lots of reading over the summer."

Not that your response has to be quite as detailed as this, but here's one to get you started:

(Original post by henryt)
I have just finished my MMath at Oxford, and as part of that I did a dissertation, which wouldn't be too dissimilar, I would've thought, from an MSc one - certainly, some of the Oxford MSc students (MSc Mathematics & Foundations of Computer Science, or MSc Applied Statistics) share courses with the 3rd and 4th year MMath students, so I don't see why the graduate dissertation would be marked with significantly different criteria. I got 73, which is a First, so I'm guessing equivalent to a Distinction...
I think I was extremely careful to be organised with the papers that I read - I used a tool called BibDesk (I use a Mac - I don't know if an equivalent version exists for Windows) which stored the pdf versions of the papers I'd read together, and either downloaded part of the citation for me, or allowed me to edit it in a form really easily, which I could then output as a BibTex bibliography. I then compiled this when I was building the LaTeX document. I'd say presenting your work in LaTeX is really important, as it is not only nice/impressive to read, but it does make the Bibliography streamlined for the reason about, and it's clearly one of the best ways of word-processing and presenting Mathematics. If you're not confident with LaTeX, use LyX - it's a good stepping stone between full blown LaTeX and a bog standard Word Processing application.
Other than that, just make sure that you do something new. It doesn't have to be insanely original or clever. Just take the work in a direction that hasn't been seen before. For my dissertation, that involved changing a proposed model from some work by some financial economists, and discussing and simulating the implications and results. Whilst I'm sure there were plenty of shortcomings in my analysis, I think that the time and effort involved in doing something original will score you marks.

As for the exact marking criteria, I'm sure your university website will have some guidance. After all, my work could be marked on and entirely different basis to yours, so it's best that you find out precisely for yourself! Good luck with it - if you put in consistent effort across the year, you'll be fine.
Hopefully we'll get a nice collection together! Thanks for your help!