Master in statistics vs operation research?
I am thinking a master in stats lately
most likely to enchant my quant background (i am finance major)
i dont know if i will go for quantitative analysis in bank or phd and i dont even care right now
i wanna do it because i like statistics aswell.
my university offers a program with 2 routes (its in the statistics department)
both have the same core
propabilities
statistical inference
linear models
time series
operation research
multivariable analysis
now, route 1 applied stats have
applied stats
General Linear Modelling
bayesian stats
quality testing
categorical analysis
biostats
advanced samplimg methods
route 2 operation research
risk management
financial math
non linear programming
logistics
dynamic programming
"waiting theory" (sorry for my english)
inventory management
i am in a dilemma, which of the 2 can i self study easier? (so do the hard in the university and the easy in my free time)
-------------------------
forgot the electives
both sides
operation research-discrete optimization
simulation
financial math
non parametrical analysis
3/4
imo you should go with statistics, it's more broadly applicable and I think most people would find bayesian stats and GLM more fun than logistics, inventory management, or linear programming
But really, just do which one sounds more interesting
As a person who studied pretty much all subjects from both lists, I can tell that both give your brain a very good exercise on a daily basis. Stats would be easier to study (financial math and risk management, if this is the same what I had, are fairly complex); however, financial math and risk management are must-have for quant, in my opinion. At least in my course, risk management provided most relevant and advanced techniques for financial market analysis, hedging and speculative strategies, all very quant-oriented.
Linear programming is definitely applicable to finance. If you want to do Asset-Liability Matching or even portfolio optimization, it's helpful to have some familiarity with linear or convex programming. I wouldn't dismiss it and if you teach it the right way- in the context of a portfolio engineering class- it can be extremely helpful.
But I think the stats will serve you well, too. Stats is always helpful for traders and sellside quants, and it's even more helpful for quant PMs reading financial papers.
I'm still a little interested in that financial mathematics class. Do they teach stochastic calculus? How deep do they go? Girsanov? Feynman-Katz? How much econometrics do they teach?
I really think you could go either way. Either way you have your work cut out for you. A master's degree in ORFE is going to involve about 1000 pages of stats, calculus, stochal, economic models, linear algebra, and optimizations (I should know- I just finished an ORFE Master's degree.). A degree in statistics could be even more work. In order to get through an MS, you need to be good at what you're doing and you need to enjoy it on some level.
Good luck and if in doubt, choose what you enjoy more. The competition for quant jobs on the street is so tough that it's more important for you to be doing and studying what you enjoy than it is for you to do what employers are hiring more people for (provided that they are hiring in sufficient quantities, which is the case for Statistics and Operations Research)
TLDR: Do what you enjoy more. Employers value both. There is no wrong answer here, so long as you're not majoring in Art History or Philosophy.
neither. Financial engineering is what you're looking for.
thanks for answers, i will just comment what i see in your answers about the financial math and risk management, i didnt go into heavy quant stuff, but i have already taken the courses (at least in theory) as a finance major. Generally i am like "dont teach me more pricing/Asset Management models, teach me how to properly estimate all those i already know"
about the linear programming i am thinking, if i take discrete optimisation, and i am smart enough to get a degree in stats, cant i just program it in VBA or something? as i said i try to get the route that i cant self study.
Anyway thank you for taking your time to answer me, really appreciated
thanks for answers, i will just comment what i see in your answers about the financial math and risk management, i didnt go into heavy quant stuff, but i have already taken the courses (at least in theory) as a finance major. Generally i am like "dont teach me more pricing/Asset Management models, teach me how to properly estimate all those i already know"
about the linear programming i am thinking, if i take discrete optimisation, and i am smart enough to get a degree in stats, cant i just program it in VBA or something? as i said i try to get the route that i cant self study.
Anyway thank you for taking your time to answer me, really appreciated
thanks for answers, i will just comment what i see in your answers about the financial math and risk management, i didnt go into heavy quant stuff, but i have already taken the courses (at least in theory) as a finance major. Generally i am like "dont teach me more pricing/Asset Management models, teach me how to properly estimate all those i already know"
about the linear programming i am thinking, if i take discrete optimisation, and i am smart enough to get a degree in stats, cant i just program it in VBA or something? as i said i try to get the route that i cant self study.
Anyway thank you for taking your time to answer me, really appreciated
I take it you don't actually know what linear programming is? http://en.wikipedia.org/wiki/Linear_programming
worth a read
Linear optimizations are often done in a linear optimization-specific language, IE AMPL. (Though you have solvers with C++ and Java APIs.) But you still need to understand the semantics and a little bit of the math and logic behind it.
You can really think of a linear program as something from calculus- finding the local maximum of a function- only though the function is multivariable with 10,000+ variables on the right hand side. At some point, calculus breaks down and you have to start using the interior point method (or the simplex method in the case of a linear problem). That's where convex nonlinear programs come in.
There's nothing wrong with doing an MS in OR or Statistics, but I do think it could be helpful to do a little more research before you jump into this. Now obviously you're already doing some of that, which is good. Part of your research has been coming to WSO to ask some questions, and we've helped you identify some of the places where you don't know what you don't know. But make sure you go down the syllabus in each of these courses- ideally with a current student there- and try to figure out what's going on.
Tenetur saepe rem asperiores consequatur. Modi velit dolorem distinctio. Est quasi vero dolores et.
See All Comments - 100% Free
WSO depends on everyone being able to pitch in when they know something. Unlock with your email and get bonus: 6 financial modeling lessons free ($199 value)
or Unlock with your social account...