I am in the first semester of a MS in mathematics. A requirement for the degree is to write a masters thesis. Here a thesis means writing on a current area of research in finance, but an original contribution is not expected (although encouraged when possible.)

There are no professors at my university who research in quantitative finance. I am allowed to write a thesis on quantitative finance, but I was told that I would have to find the "problem" on my own to research.

So my question is, how can one find an area of quantitative finance or problem to study that is appropriate at this level to write a thesis on?

Also, how does one know what has already been "done"? For instance, I was reading about Fast Fourier Transforms and their relation to finance. I was considering extending the NVIDIA CUDA Fast Fourier Transform Library for finance applications. But I imagine this and many other things have already been tackled already.


I can tell you what I did for my undergraduate thesis (which was a required component on my degree in Mathematics.)

I picked a paper on option pricing which gave, as a main result, the closed for solution for the price of an American and Asian Put Option. My contribution was to apply the Fourier Transform to the initial equation in order to simplify the problem and then use the Method of Characteristics to get a solution. This way I could compare my solution to theirs and figure out whether I was right or not.

In essence, I took an existing and already "solved" problem and decided to tackle it in a different way (by using the Fourier Transform) and using a different technique to solve it (the Method of Characteristics.) Maybe this kind of approach could work for you as well.

Hope this helps and good luck!


I am at the beginning of my graduate research in the area of finance, and have a Bachelor's degree in mathematics. A first approach would be to look for the right keywords in databases, such as Jstor or SSRN, but it may be a bit difficult to find quantitative topics in the latter at least.

It is not necessary to first look for what has been done. If the paper needs not to be original (though encouraged), replicating other results with new datasets or different methods (as g_puffo suggested) is also a valid option.

You should first comb through current literature and see which topic or field of research interests you more than others, and a fitting topic suitable for a thesis will arrive by itself.


Not the answer you're looking for? Browse other questions tagged or ask your own question.