Suggestions for topics for master thesis

Rank: Chimp | 9

Hello!

I have one year left of a Master of Finance degree. Over the last two quarters (January-June) I have to write a thesis. I want to work in IB after i graduate so I want to find an IB-related topic. Valuing a company is considered a "standard" topic on my school, and it is often easier to get better grades with more "innovative" topics. Does anyone have any suggestions for a topic that is "innovative" and also prepares me for an investment banking job?

Thank you!

Comments (19)

 
8/27/13

EDITED to give OP an IBD project for a Master's student.

How comfortable are you with linear programs, stats, and stochastic optimization?

Maybe you take a look at resource companies. One anomaly with gold miners is that they mine their most costly gold during periods of high prices and mine their least costly gold during periods of low prices.

Assume gold is a stochastic process that has drift of inflation and standard deviation of returns equal to its historical average.

Take a look at a specific miner's mining opportunities and:

1.) Figure out why it wants to mine the most expensive gold first rather than go for the least expensive gold. This is where the stochastic program/optimization comes in. You could also do it discretely and just use a markov decision process and work your way backwards. But there will be some coding.

2.) Is this efficient management or is this a conflict of interest? Are mining companies operating in the best interests of shareholders by preserving their options to mine later? Or should they be exercising the cheaper effective calls from the less expensive mines today? What are the decisions coming out of the optimization recommending?

3.) If it's efficient management, figure out a product to sell that takes advantage of the optionality of this gold mining.

4.) If it's a conflict of interest, focus on which companies are being poorly run based on the model developed and the potential value of a takeover.

You still get the valuation perspective from this, but now you have just developed a really innovative tool and explained an interesting phenomenon in business. Someone may have already looked at this and published it, but otherwise this could also be publishable. Not a top three journal but maybe a 2nd tier? Either way, it's going to make the professor reviewing it think long and hard.

If you can get this published in a 2nd tier journal, you've pretty much got a call option on a top 20 finance or econ PhD if you hate IBD.

I know this stuff may sound really complicated, but all you're doing is modeling gold prices on a normal distribution and treating a series of mines like options in a binary pricing tree. And you may be creating a new way to value gold companies.

Maybe someone has already done this. And I think most analysts are aware that on some level gold mines are options. But I don't think the ER mining groups are running stochastic models to do their valuations, and it's possible the "mine your most expensive gold first" phenomenon is still an unanswered question that a stochastic model could help explain.

Learn More

7,548 questions across 469 investment banks. The WSO Investment Banking Interview Prep Course has everything you'll ever need to start your career on Wall Street. Technical, Behavioral and Networking Courses + 2 Bonus Modules. Learn more.

 
8/27/13

^^this is gold (no pun intended)

 
8/27/13

I would also point out that there may be some phenomenon explaining this that the simple model I outlined may not explain. You may find something very counter-intuitive. There may be startup or shutdown costs. The price of gold may not follow a normal distribution and/or you may be dealing with a markov-switching volatility model with periods of high and low volatility. It may just be that management is running the company in their own best interests (self-preservation). You'll probably see this as a wrench in your research project at first, but it may be an opportunity to prove something interesting about how mining companies (or even businesses in general) operate. Depending on how much data the explanation requires, you can start taking a look at other companies, see if they have the same problem or not, and ultimately try to develop some sort of strong t-stat proving the correlation between the deviation you observe and some sort of underlying condition/explanation about the company.

So I would try to find any sort of literature or information I could on this market behavior, take a look at a firm that publishes a lot of information on mining costs and which mines are operating, and see how "rationally" they are operating relative to the suggestions of the stochastic program.

Also, I know this post isn't on Arxiv, but it's timestamped on WallStreetOasis, so it's going to be embarrassing to claim sole credit in a major journal. If this brilliant idea gets published in a top three journal, I request a piglet, boar, or other porcine creature to be shipped to me via air freight (cannot be ground- MUST BE FLOWN) and a footnote acknowledgement for "IlliniProgrammer" in the publication.

I do think you'll find something interesting (it may be a lot of work).

 
8/27/13

I'm willing to help a little with explaining how to set this up if OP knows how to code ok in a statically typed object oriented language (C++, Java, etc.) and can understand the basics of how gold prices move (the variance is proportional to time; the standard deviation is proportional to the square root of time).

You are probably going to have to optimize on a monte-carlo with millions of simulations to get a general strategy to converge, so you can't write shitty code for this. It doesn't have to be great, but you can't be doing O(n^2) stuff.

 
1/2/15

Hi, I'm currently looking for a interesting topic for my Master thesis, and when I was goggling around i stumble over this discussion. This is exactly what I want to do so I was just wondering if you are still as excited about the subject and if you know if someone already wrote about it?

Best regards,
Gustaf

Gus

 
9/24/13

I apologize for the unprofessionally late reply (I've been on a roadtrip with restricted internet access). Wow, that was a great suggestion. Unfortunately I don't know enough about neither gold mining nor coding to write about this. Hopefully someone else can make use of your tips! Thank you IlliniProgrammer!

 
9/24/13

I'm also also in the exact same boat. Could of course do a (standard) valuation exercise about Company X, but that's as original as buying an iPhone.

 
1/2/15

Sometimes I think I am smart but then I read an IP post

 
 
1/2/15

the average price of a phallic

 
1/2/15

Risk hedging in commodities markets.

 
 
1/2/15

I'm currently researching the implied base correlations of the cdx index tranches during and post the financial crisis. Definitely do something fairly quantitative it helps a lot in interviews

 
1/2/15

Look up Darrell Duffie and his work on Dark Markets. Pretty interesting stuff. Maybe you can find some inspiration there.

-MBP

 
1/2/15

Implementation of a negative income tax on total income tax revenues over the last 40 years as compared to the current system

If I had asked people what they wanted, they would have said faster horses - Henry Ford

 
1/2/15

Before I delve into Duffie, would that be the research on incomplete markets in equilibrium?

"The only point in making money is so you can tell some big shot where to go."
-Bogie

 
 
1/2/15

- m&a wave?
- emergence of boutqiues (along the line of stanford business school academics like Michael Hannan) in ibd
- m&a in fish markets (like processed ones)
- demystifying investment processes in a big Asset Management firm or team (a lot of BS as they tell you)
- airline industries (corporate finance, revenues, debt etc.)
- restaurant business and changes in margins,
- how would a research team survive in the banking in the future?
- is there money in wall street?
- google, what exactly do they want to do with their company?
- lies in banking: diversity discussion!
- racism in banking?
- sexism in banking?
- what does the hr actually do in banking?
- can we accurately forecast the dividend growth rate (and if you can, send it to Chicago Booth)?
- real growth or just facade? why there is no income increase despite the claim of the economic pick-up?
- job figures? are they telling the true story? 3000 jobs created at p&g, yeah but they receive million applications every year or quarter?
- why dont we outsource IT to India?
- why dont we move non-client facing roles to countryside or even to a boiler room and pay those who do crap (despite the fact that we can)?
- why dont we hire more into ibd if one employee does 120 hours a week? we can hire 2 and each does 60 hours and they split the cheque? inefficient hiring practice in banking?
- what do traders do now in banking? i know one girl who joined credit swiss in 99 and now an anchorwoman at bloomberg.
- obama, what does he do all day?

just dont write crap.

 
4/26/18

1-Click to Unlock All Comments - 100% FREE

Why do I need to be signed in?
WSO is a knowledge-sharing community that depends on everyone being able to pitch in when they know something.
+ Bonus: 6 Free Financial Modeling Lessons with 1-Click Signup ($199 value)