I wish to write my thesis in quantitative finance, but as I understand it, a thesis needs to have some sort of originality to it. You can't just take some theory written by others and just rehash it in your own words. You need to do something never done before.

How does one accomplish this in finance?

Because, in my head, when trying to come up with some ideas for a thesis, they all fall within the same structure of

  1. Introduce some problem (say, a pricing problem)
  2. Introduce methods to solve the problem .
  3. Apply the methods to some specific setting and compare.

Now, in step 1 and 2, you aren't doing anything original, indeed this part of your thesis is literally just an exploration of existing theory (unless you come up with your own theory, which is kind of unlikely in a master's thesis given you only have 3-4 months to do it).

In step 3, you can do something original by choosing a setting or numerical application which is not seen before (e.g. you can apply your methods to some index that it hasn't been applied to before, compare different countries or time periods, or something like that). And of course the fact that you are comparing a specific set of different methods may not have been done directly before either.

But that doesn't seem like a whole lot of originality.

So please, can anybody enligthen me on how one ensures that a master's thesis does something original and is not just an exposition of theory written by others?

  • $\begingroup$ "the doesn't seem like a whole lot of originality" might be part of your problem. progress most research fields is incremental for the most part, steps versus jumps. particularly for a thesis, even moreso at a Master's level, I'd expect 'originality' to take the form of precisely that kind of 'step originality'; that, or doing some kind of work that adds value but might not be the 'sexiest' problem. $\endgroup$ – Chris Jun 11 at 23:06
  • $\begingroup$ I'm sorry, but it's very unlikely you'll produce something original, effective and correct at the same time: in this field, we've been stuck for 30 years around slight variations of concepts belonging to early XX century and quantitative finance usually steals tools and concepts from other fields before giving them high-sounding names to hide the theft. I think that the best trade-off would be to put together some papers that are only theoretical, implement code, and test/analyze concrete results. Then disguise behind a title like "An empirical approach to..." or other old verbal patterns. $\endgroup$ – Lisa Ann Jun 12 at 8:40
  • $\begingroup$ Agree with Lisa Ann, it's going to be quite difficult for you to produce something really original (but not impossible of course). Probably best to either choose a topic you really like and add small variations to it that you can handle within the given amount of time, or jump on the Machine Learning bandwagon and do whatever it is ML does to give you a higher chance of landing a job after your thesis. $\endgroup$ – ilovevolatility Jun 12 at 10:48

You can choose an empirical issue. Given the short period available to complete your Master Thesis, introducing a new pricing method and do something never done before is quite challenging. However you can use your knowledge to examine an empirical issue. For instance, the hedging efficiency for a certain product, the connection of credit risk with firm specific variables or macroeconomics. The list of possible topics is neverending. Indicativelly, you can have a quick look on the Journal of Empirical Finance, Quantitative Finance, Applied Financial Economics or any other Applied-Finance oriented journal.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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