Serving the Quantitative Finance Community

 
User avatar
bos09
Topic Author
Posts: 0
Joined: September 26th, 2008, 11:19 pm

I need a topic for my thesis...

January 3rd, 2009, 12:08 am

Hey guys, so I'm in my senior year of undergraduate study and am doing a half-year thesis on financial mathematics/financial engineering or anything related (even pure probability). I have a strong interest in probability and would like to do something largely theoretical, but at the same time I want to avoid writing a theoretical paper based on a lot of completely invalid assumptions about the markets. I'm not sure if these two things contradict each other, but I am very well open to suggestions. I was hoping to do something related to arbitrage and/or market microstructure, but again, I am really open to any ideas (I was even thinking of creating an efficient poker system). The main problem is that I'm not sure what problems are out there and what has already been covered, so any direction in this light would be greatly appreciated as well. Thanks guys
Last edited by bos09 on January 2nd, 2009, 11:00 pm, edited 1 time in total.
 
User avatar
ExSan
Posts: 493
Joined: April 12th, 2003, 10:40 am

I need a topic for my thesis...

January 3rd, 2009, 1:52 pm

Madoff ?
°°° About ExSan bit.ly/3U5bIdq °°°
 
User avatar
Alan
Posts: 2958
Joined: December 19th, 2001, 4:01 am
Location: California
Contact:

I need a topic for my thesis...

January 3rd, 2009, 5:45 pm

QuoteOriginally posted by: bos09 I was hoping to do something related to arbitrage and/or market microstructureA current question in the Student forum gave me this idea. I suggest some kind of survey relating to put/call parity and its violations in the listed options market.This is one of the most basic arbitrage relations -- it gets violated when stocks become 'hard to borrow'.You could survey: (i) existing studies; (ii) famous hard-to-borrow situations (Palm computer and Porsche/Volkswagon cometo mind immediately); (iii) how does the borrowing market really work; (iv) after you know your subject,try to add something interesting/original.If you can achieve point (iv), post your thesis here or maybe even submit it to Wilmott mag. Good luck,
Last edited by Alan on January 2nd, 2009, 11:00 pm, edited 1 time in total.
 
User avatar
mit
Posts: 0
Joined: February 5th, 2005, 4:52 pm

I need a topic for my thesis...

January 6th, 2009, 9:31 am

do one on subprime derivatives pricing in the post-subprime era
 
User avatar
niki5
Posts: 1
Joined: January 25th, 2007, 8:10 am

I need a topic for my thesis...

January 6th, 2009, 11:28 am

Pricing of Moment Swaps and their practical implementation
 
User avatar
bos09
Topic Author
Posts: 0
Joined: September 26th, 2008, 11:19 pm

I need a topic for my thesis...

January 8th, 2009, 12:49 am

Hey, so thanks for the replies guys.robsan - ??Alan - That doesn't sound too interesting to me; it seems like it'll be pure data processing and quite boring to me. I'm trying to avoid the tedium of data process as much as I can, especially since I can choose almost anything to do a thesis on.mit and niki5 - those seem like ideas that I might use, but I don't know, for some reason it just doesn't make my body tingle.One of the things that I'm strongly considering is along the lines of agent-based modeling, irrationality, and market-microstructure. Namely, I want to see what kind of results you get if you assume that not all agents are rational. In particular, if there are some watershed results in the field that assume rational-acting agents, see how the results are changed when you throw in a portion of irrational agents - see if it is very sensitive to the rationality assumption; and if it is, point out that it's not entirely prudent to be using the model since we know quite well that agents aren't rational in the real market. Alternately if there aren't really any landmark, perhaps try to create some situation, maybe even outside of financial markets (again, my topic doesn't have to be strictly in finance) and instead in game theory, and then again, see where it's sensitive to irrationality. I know Doyne Farmer has written some papers on zero-intelligence agents in financial markets and read a bit on it, but not sure how important these are.If anyone can give me any leads in this area, that would be awesome. Thanks a lot guys.
Last edited by bos09 on January 7th, 2009, 11:00 pm, edited 1 time in total.
 
User avatar
C3I2
Posts: 0
Joined: November 26th, 2003, 12:58 am

I need a topic for my thesis...

January 8th, 2009, 10:41 am

Why not look at games, instead of Financial markets. Betting markets have interesting qualities that help you look at behaviour. Also, now there is bound to be lots and lots of data, both on an aggregate level (just Googe: paramutual betting markets for examples of studies) and individual data. This means that if a company give you access you can study individual behaviour.
 
User avatar
Alan
Posts: 2958
Joined: December 19th, 2001, 4:01 am
Location: California
Contact:

I need a topic for my thesis...

January 8th, 2009, 5:09 pm

Sorry you didn't like my suggestion, but I took you seriously when you said:"I want to avoid writing a theoretical paper based on a lot of completely invalid assumptions about the markets."Anyway, I am not trying to pick a fight: just wanted to add that no data processing was involved in my suggestion.I can't make it less boring though, except to say that if you get involved with options, and come back in 10 years,it might look a little better.
Last edited by Alan on January 7th, 2009, 11:00 pm, edited 1 time in total.
 
User avatar
bos09
Topic Author
Posts: 0
Joined: September 26th, 2008, 11:19 pm

I need a topic for my thesis...

January 8th, 2009, 5:19 pm

QuoteOriginally posted by: AlanSorry you didn't like my suggestion, but I took you seriously when you said:"I want to avoid writing a theoretical paper based on a lot of completely invalid assumptions about the markets."Yeah, that's kind of my point. Of course I will be making assumptions when I suppose that some certain proportion of participants act irrationally, in a specifically irrational way, but I'm not meaning to do that to get a result. Rather, I want to show that results that have ALREADY been obtained based on a silly concept of perfectly rational agents... are not robust if that assumption is violated. I'm not sure how useful this would be, or how useful anything that could come out of it would be... but that's what I was thinking about.QuoteAnyway, I am not trying to pick a fight: just wanted to point out that no data processing was involved in my suggestion.Sorry, just re-read your post. I thought you'd meant that I should go data sifting through options prices and seeing when they don't satisfy put-call parity - this would be boring as all hell. But now I think you're meaning to say that I will be able to find good examples of it happening without sifting through data (e.g. I will find it in other papers or something), and I should do research on what I find? I'm not sure I see where theoretical work or probability comes in - that's really my main interest. I think it'd be hard for me to foment any effort into the thesis without tossing some probability on it.
Last edited by bos09 on January 7th, 2009, 11:00 pm, edited 1 time in total.
 
User avatar
Alan
Posts: 2958
Joined: December 19th, 2001, 4:01 am
Location: California
Contact:

I need a topic for my thesis...

January 8th, 2009, 5:36 pm

Yes, you will find lots of already-done analysis in the literature. Whether or not there issomething you can add to it, involving theory or applied probability or whatever, who knows?You have to read the existing literature first and then have a good idea.