DISCUSSION on second rated MFE programs

[bigbadwolf]

I see Dominic Connor has made some perceptive and disparaging comments about Fordham's MSQF program at his Wilmott blog. He's doing us all a service. It would be nice if people could contribute the names of similar garbage programs. There's a plethora of such, as ill-equipped institutions frantically attempt to cash in on the bonanza of quant education. Yet the students these institutions ensnare -- poor suckers -- are ill-served by the "programs" offered (I use the term loosely): the hollowness of their training will be exposed within five minutes at interview stage (if they get that far). As I always maintain, forewarned is forearmed. Somewhere in cyberspace there should be a place where critical reviews of various MFE programs are available.

 

[Tigga]

I read the whole Fordham episode on WM. I believe it came about first on GD as a current student there got pissed off about the quality of Fordham program and vented on GD, WM.
Anyhoo, my understanding from reading the thread is that Fordham being new kid on the block and promised more than it can deliver. The student paid 40K and didn't learn C++ or get an internship.
You can read that thread on WM here
I'm not associated with any program but from the readings the last few years on the forums, there seems to be a more business-like approach from many new programs.

It appears to me now that many university opens competing programs ? Columbia has like 4,5 of them ? I also read stories that if you apply to Columbia MFE and didn't get in, they will just conveniently "recommend" you to their MSOR program or whatever there. It's funny to see how many naive students being a sucker and go for it. They got to keep the suckers inside the system and paying through the nose for second class treatment.

A while ago, I also heard that Poly program was really bad. When some student went to their open house and asked how Poly would help them get a job, the director asked him whether he currently working. Yes, he said and the director replied : "There you go. Problem solved". :wall
To their credit, they have tried to improve the program since then. I heard they got a new director and the current students seem to be happy with their efforts to repair the bad rep.

I'm not trying to blame any program. Just want to make sure students really do their research and not get sucked in. There are too much money involved for one to go to a program for all the wrong reasons.

 

[MikeK]

Rutgers also has two competing programs--the MS Math Finance from their Arts & Sciences school and the Quant Finance program in their business school. Once you get past the obviously elite programs like NYU, Baruch, CMU, etc, it is hard to differentiate between bad programs and programs that are new but show potential. In my opinion--and I'm sure most agree--the best gauge is how well the graduates fair in the job market. And this probably has a lot to do with how good of a job the program director does in making the program visible to the financial world.

I'm guessing there are a good number of new programs out there that figure the nature of the FE degree in itself will guarantee students get good jobs, and they do little to push it any further.

 

[Tigga]

MikeK,
Wholeheartedly agree with your points. The effort by the head of the program is very important. I read the thread on WM about Fordham and here is what DCFC posted

As a HH, I have data on which schools I see as good at helping their students get jobs. I don't have numbers it's a "good, mediocre, crap, or unknown" scale.
Baruch is very good, as is Courant, Imperial, and Cass is OK(though I get some bitching about Cass, I think they're rather better than average).
People like Dan Stefanica of Baruch, and Peter Carr of Courant do a fine job of making sure their kids get placed.

I'm amazed how much the perception of the Baruch program changed. If you compare Baruch with NYU, Columbia 2 years ago, a lot of students will laugh and said Baruch is in 2nd or even 3rd tier. I guess they aren't laughing now. While some may still disagree, most will accept that Baruch is a quality program and fully committed to improve its program by leaf and bound. Baruch will never has a name brand cachet but it is clearly doing all the right things.

At the end of the day, people go to these kind of programs to get a job and how well the students perform in this kind of market and how committed the director to the program make a whole lot different.

A lot of programs rely too heavy on their name and pay little effort going out, making connection with headhunter, hiring managers, etc.

No matter how glossy your printed ads are, your program is a product judged by the students and hiring companies. They are the customers and the customers are always right.

 

[bigbadwolf]

Rutgers also has two competing programs--the MS Math Finance from their Arts & Sciences school and the Quant Finance program in their business school. Once you get past the obviously elite programs like NYU, Baruch, CMU, etc, it is hard to differentiate between bad programs and programs that are new but show potential. In my opinion--and I'm sure most agree--the best gauge is how well the graduates fair in the job market. And this probably has a lot to do with how good of a job the program director does in making the program visible to the financial world.

I'm guessing there are a good number of new programs out there that figure the nature of the FE degree in itself will guarantee students get good jobs, and they do little to push it any further.

All this talk of "elite programs" unsettles me. In the realms of mechanical engineering and electrical engineering, we don't place so much emphasis on what program is or is not "elite" (though clearly there's a pecking order, with schools like MIT and CalTech, up there at the top). In financial engieering it seems to be more of a make-or-break situation: if you have an MFE from Haas or CMU, the odds are vastly more in your favor than if you have an MFE from Swampwater State U. What one would expect from a decent program, however, are two things:

1) Decent courses, taught by academics who know how to teach, and where one is properly exposed to Monte Carlo, C++, numerical analysis, stochastic calculus, etc over a period of time. Thus, for example, Monte Carlo cannot be covered in two lectures; it requires at least a semester, and using a decent text like Fishman or Robert and Casella. Stochastic needs two semesters with something like Shreve's two volumes. And so on.

2) Active placement programs, where the FE department has close liaisons with industry, and can place people in internships and regular jobs.

Some of these half-baked half-assed programs have neither holding. The instructors often don't know the material themselves. The situation resembles what happened on American campuses during the sixties, when there was a growing demand for computing degrees but the universities didn't have the expertise to teach them. Some "programs" have profs who know virtually no finance and/or who hire adjuncts ("industry experts") off the street. What a pathetic joke.

And in universities not in proximity to the main financial hubs -- NYC, Chicago, San Francisco -- they can't find internships or jobs for their grads. Usually they don't even have working relationships with industry.

 

[bigbadwolf]

I'm not trying to blame any program. Just want to make sure students really do their research and not get sucked in. There are too much money involved for one to go to a program for all the wrong reasons.

Easier said than done. All the schools offer glossy brochures that are at least partly designed to camouflage sordid realities (their teaching may be garbage or their placement efforts pathetic). That's why there should be some place -- Quantnet? -- where critical and objective appraisals are available.

If a company is interviewing a mechancial or electrical engineer, it expects a certain toolkit of skills and knowledge from freshly graduated engineers. The same can't yet be said of MFEs because of a wide variety in quality. There have to be standards. And there won't be standards without a continuous process of criticism, appraisal, and accountability.

 

[Tigga]

All this talk of "elite programs" unsettles me.

Do you know that some program even uses the "ranking" on GD to tout their program on their website ?

The Haas School's Master's in Financial Engineering Program (MFE) ranked #1 among financial engineering and quantitative finance master's programs around the world, according to Global Derivatives, an information clearinghouse for quantitative finance topics.

How lame can this be ?
I mean it's 2008 now and we all know "Global Derivatives, an information clearinghouse for quantitative finance topics" is a joke and as is their ranking.
UCB is an excellent program, if not one of the top but using thoroughly outdated ranking with questionable criteria just goes to show you that how much people willing to gloss over details and use whatever means to promote their products.

Now, for someone from Asia who knows no better about the US education system, seeing something like #1 ranking on the world according to GD, who blames them for sucking it up :D
After all, GD is the authority on everything FE related.

 

[DominiConnor]

I've got shot at for this, a couple of people claiming that it's some spite against Jesuits.
Yes, really.

I have only partial info the quality of the Fordham course, odd but bad things like accountancy and VBA instead of C++. That is not uniquely bad, and would have merited only a passing comment.

But I found that they made serious and clear promises of internships. As many around here will know, IB internships are tough to get, indeed it would be rational to pay the full $40,000 just to Fordham to get a top tier internship. Some firms explicitly target other firm's interns (not small ones either, big brutes).

They continue to make these promises, it's on their website right now. The word "capstone " is used.Fordham Graduate Business :: Fordham University

Their presentation also promises an internship:
http://www3.bnet.fordham.edu/files/MSQF-Presentation120507.pdf

It also has an impressive list of Internship Providers

•Bear Stearns
•Goldman Sachs
•Credit Suisse
•JP Morgan Chase
•Merrill Lynch
•Orca Asset Management

I'd read that as being promised serious goodies, hands up those here who turn that lot down ?

My information is that this has not happened big time. An excuse has been that the "current market" has done for them. I reject that utterly. Setting this sort of thing up takes at least a year, which means long before the current mess. They relied upon the market staying buoyant with no hedge, an error that would fail a newbie quant at interview, let alone people that are supposed to be teaching it.

I do not get any impression that they are putting in the necessary work to fix it, and it is a lot of work.

Baruch's image is improving through hard work. When my partner Paul Wilmott (the P in P&D) started the Oxford FinMath programme, it was the 2nd in Europe. Now it's a huge competitive industry. Fordham seems to have missed this.
Indeed ,they seem to have missed what is going on, it's not exactly low profile, which tells me a lot about their awareness.

Wolf is right that it is awfully hard to get good objective information, the brochures are so similar that you could swap the contents of any two of them without noticing.

I have some info on relative quality, but it's far short of a full analysis, which is why I so rarely throw my toys around like this.
But when I see the crap that is put around about "we are a top school because Facebook says so", or whatever I am tempted.

 

[charlieLuo]

DominiConnor ,

Would you kindly provide us a list or a partial list of FE programs that ring a bell at big name employers?

Thank you very much! and I believe most quantneter here will do as well.

 

[Andy Nguyen]

This thread has a potential of being a flame mine field or a possible constructive great read. We don't want to use this thread or Quantnet to badmouth any program but we want to point out flaws in the way prospective students approach these programs.
I'm from Baruch so I'm biased by association but I'll add a few comments.

One, there is a big void of unbiased info about these programs, especially those second-tier programs or what have you. The only info you get is from their website. PERIOD.
So when students go look for unbiased info they run into a bunch of clueless people who unintentionally pass around the same kind of wrong information. This goes on and one and bad info becomes truth, clueless generation X becomes expert giving advice to clueless generation Y.
There is no better example about this than what is going on at GD. To be fair there are few helpful, knowledgeable people but they are hopelessly outnumbered and eventually have long left the building. Its funny that Tigga mentioned people use GD ranking for marketing material.

Secondly, the way these students go about picking a program and potentially their career. I have seen questions like "I got admitted into X,Y,Z. Anyone knows what the job, internship situation is like at X,Y,Z ?"
I mean how did they decide to apply at X,Y,Z in the first place. Seriously.

We are in the internet age and people should have learned to use the tools by now. These are the same group of people who spends days, weeks using internet to research on a computer part, camera, laptop that costs a few hundred dollars. And they can't do a proper research on something they are about to drop 60 grands on ?
Email, phone, alumni, current student contact should be used to find out everything they can. Not every program has a quantnet portal like Baruch but it's not an excuse.

If people refuse to give you important info like internship, job or allow you to contact alumni, current students it should give you pause. If they don't want to help you when you want to give them 60K, how eager they will be once they took your money.

Now if you add point one and two, you get a picture of what going on. Good and unbiased info is just so hard to come by. The quantnet wiki is an effort to give students a place to add and edit info relating to mfe program. I think I emailed the directors of Fordham program a while ago but they never bother to edit their wiki entry.
Incidentally, the quantnet wiki came about when Dan Stefanica of Baruch and Peter Carr of NYU decided to put together a place to share all info about mfe programs in NYC. They know there aren't much info around.

The Fordham incident reminds me of someone on quantnet last year who got in fordham but rejected at Baruch. He asked me if he should go for it. Knowing nothing about Fordham at that time, I told him to improve his profile and try again at Baruch. He did.

Imagine what may happen had I told him to join Fordham. He probably got so pissed he may hire some hitmen to kill me to revenge the material and mental loss he suffers.

It may be one of the best advice I gave on quantnet.

 

[charlieLuo]

The first tier programs


I think it makes a lot more sense to figure out "the first tier programs"

others are those who do not have good placement. if so it does not worth the $$$$$$. so simply avoid em. no need to learn more about them. if can not get into a first tier, do something else maybe is a better idea. this is similar to MBAs. anything 2nd tier or lower has much lower value than the first tier.

also, 2nd tier FE program can be derived from the list of first tier.

so, I'd suggest that we focus our energy on the first tiers, and stay away from sad stories/rumors etc.

I might be wrong though. But I am pretty positive on this.

 

[Andy Nguyen]

I think it makes a lot more sense to figure out "the first tier programs"

others are those who do not have good placement. if so it does not worth the $$$$$$. so simply avoid em. no need to learn more about them.

Some have tried. None has succeeded.
If you really think about it you will realize that you are doing a fruitless exercise.

If we go by placement rate, we would have ignore those programs that not make the placement stats public. That will eliminate Columbia MFE, Stanford, etc...

That will leave UCB, CMU and Baruch right at the top 3 programs with Baruch having close to 100% rate the last few years.

If you look at the cost of education, then Baruch is undisputed #1 with a cost of 9K for NY residents and 18K for everyone else. CMU comes in last at close to 100K tuition.

If you calculate the ROI by dividing the first year average salary for students with no experience to the cost of education, Baruch comes out something like 90K/18K, a ratio of 5. If you take into account people like me who paid only 9K, the ratio is like 10.

Now for CMU to have a ratio of 5, 10, their students have to make 500K, 1M first year. How are they gonna compete with Baruch if we calculate that way.

I can twist all the calculation to have Baruch comes out on the top 5 or so but who will take my calculation seriously. Zero.

This just goes to show you that trying to rank is a waste of time. Whatever scientific formula you derive, people will always find ways to discredit it. And who will take your ranking seriously ? Like the way they take GD ranking seriously :D

 

[bigbadwolf]

The first tier programs

I think it makes a lot more sense to figure out "the first tier programs"

others are those who do not have good placement. if so it does not worth the $$$$$$. so simply avoid em. no need to learn more about them. if can not get into a first tier, do something else maybe is a better idea. this is similar to MBAs. anything 2nd tier or lower has much lower value than the first tier.

If everyone concentrates on the so-called "first-tier," then everyone will be trying to get into a handful of schools: CMU, Chicago, Haas, and two or three others. And most people will end up disappointed because the odds of acceptance will be stacked against them. And there's probably a bigger market for quants than what a handful of schools can provide. So I'm not sure that just concentrating on top-tier schools is necessarily the best advice. Allow me to elaborate.

An MBA applicant may not be able to get into Harvard or Wharton (and indeed may not even want to), but he or she can get a sound MBA education at any one of thirty or more US business schools, each of whom has established programs, with competent and seasoned faculty. This doesn't hold for MFE programs, at least not yet.

Once an MBA student graduates, he or she may not get the high-flying corporate job that a Wharton or Stanford graduate may aspire to -- but again he or she may not want that kind of job. The MBA market is large and diverse enough to accommodate several tens of thousands of graduates every year. The same may one day be true of MFE programs.

Furthermore, a prospective MBA student has available a plethora of relevant and (mostly) accurate information so as to assess the quality of the various programs, and to be able to find the best fit for himself. This doesn't hold (yet) for MFE programs.

The above paragraphs suggest we need to look at more than just a few programs and more than just a few hundred jobs in financial hubs like NYC and London. And by implication, we need to look at more than just placement with highly prestigious companies.

Placement is clearly crucially important, as it can serve as the launchpad for a career. I do feel, however, that too much emphasis is being placed on it. People also need to look at the inherent quality of the education they receive, and comment critically on it. I don't think it serves a student well in the long term if he or she is placed with (say) a cutting-edge IB team but receives only a second-class training. And as I've suggested above, a student may want quality quant training without aspiring to megabuck jobs at top-tier firms. He or she still needs information on quality programs which may not have stellar placement track records but which do a good job with regard to training.

We do need a service that provides critical information on MFE programs both with regard to placement and with regard to quality of training imparted.

 

[charlieLuo]

let me prove: NYU FE > Columbia FE > CMU/Berkley > Baruch FE > Fordham FE

Andy, bigbadwolf, and all others who hasvethought about this, I do not disagree with either of you. I think we are basically propsoing some sort of solution to a problem all of us sincerely want to crack. anyhow. let me propose a formular

quality of a FE program = f(the brand of the university, the strength of the department (and/or faculties), the location of the program, the historic performance of the program)


if we use a simple linear combination of these four factors, then it is

quality of a FE program = a1 * the brand of the university + a2 * the strength of the department (and/or faculties) + a3 * the location of the program + a4 * the historic performance of the program)

all ai's are non-negative real numbers

NYU, probably scores well on all four factors, so does columbia (I assume)

CMU and Berkeley probably score well on three of them, except location (remote thing I am not sure yet, so let me put this way for now).

Baruch scores on location and performance (according to what I heard)


now we can give a partial rannking of these schools, since all ai's are non-negative real numbers, we have,

NYU FE > Columbia FE > CMU/Berkley > Baruch FE



of course, we can further break performance down to education quality, and placement. we can do that sometime later.


if we define a program as first tier as which scores well on all above dimensions, then they are NYU and Columbia. no others. well, you might have some other idea of defining first tier though.






thoughts?

 

[Tigga]

But I found that they made serious and clear promises of internships.

They continue to make these promises, it's on their website right now.

Dominic,
You oughta watch the video presentation of their information session
Fordham Graduate Business :: Fordham University

Part 2 and 3 are full of nuggets :D
About 2/3 of part 2, the director showed a slide with Internship info on it and he pointed to the list of firms and said here are the firms we have arranged (or in the process of) internship.
He also mentioned that they talked to about 20 firms and hedge funds to find out what they want from a quantitative finance. They build their program around the list of those employers wish list. And there are no C++, just Excel and Matlab on that list. Apparently, employers do not like C++
Watch part 1 if you like. There are lot of big words flying around. In part 3, the last bullet point on the slide "Why Fordham MSQF" is Internship.

Gotta love it.

EDIT: there are more
Press release

Our new MSQF Program is market-driven. We interviewed 20 leading firms in developing the program.

The Fordham MSQF program will graduate professionals with the quantitative finance skills to value securities, develop and price derivative instruments, manage portfolios using the latest techniques, and manage the wide variety of risks to which financial firms are regularly exposed. An integral part of the program is a full-term internship for its students at leading institutions in the New York City-based financial services industry. Students accepted for MSQF will have undergraduate degrees in quantitative disciplines, such as mathematics, physics, engineering, or economics.

 

[DominiConnor]

Andy is right that students do not do enough research on MFE programs, but I've tried to research some and found no real data at all. We have looked at the idea of some sort of poll for rating MFEs, but that has problems. The biggest is that almost no one has done more than one, so comparison is hard; but also that some don't like their course because they found it "too hard". As a headhunter, I like hard courses... But it's not easy to separate tough material from bad teaching.

I would also like to provide an excuse why we don't publish a full list of MFE programs and their quality. As headhunters what we actually care about is output, not value add.
By that I mean that some programs attract smarter people, or are in such demand that they can pick the best. It is hard to accurately separate that from good teaching. A college that filters hard "adds value" merely by being a quality brand.

Thus our utility function is different from those researching MFE programs.

I also get some feedback from lecturers on several programmes (under Chatham House), that the cost of a MFE $40-50K + living in a big city for a year + lost earnings >= 100,000
means that "rich kids" have become an uncomfortable % of the students.

Aside from the inevitable lowering of quality when you select by parental wealth, rather than examination, I hear quite amazing tales of lecturers struggling to get the kids to shut up and be quiet enough to hear the lecturer, even seen it myself when dropping in on MFE lectures.

 

[bigbadwolf]

But it's not easy to separate tough material from bad teaching.

By that I mean that some programs attract smarter people, or are in such demand that they can pick the best. It is hard to accurately separate that from good teaching. A college that filters hard "adds value" merely by being a quality brand.

These remarks reflect unfortunate realities. People doing the hiring do have a different utility function than students. I just have a couple of quick remarks to make en passant. The first is there's no clear line of demarcation between tough material and bad teaching: a competent -- if not gifted -- teacher can make otherwise abstruse material clearly understandable. Indeed, this can be seen even in the treatment of material in different texts. A Shreve or a Benninga explains matters in a clear and coherent manner whereas a Meucci talks utter rot and further obfuscates what would otherwise be straightforward material.

The second is that the filters being used by so-called "top" universities -- and accepted unquestioningly by headhunters and employers -- may not necessarily be "optimal" in any sense. A potentially gifted quant may not be able to earn a math GRE score in the 98th or 99th percentile (say). Other equally essential qualities are not being tested for as part of the "filter." Conversely, someone achieving at the 99th percentile level -- and let's suppose the 98th percentile is the cutoff level -- may not necessarily be any more competent than someone achieving at the 95th or even 90th percentile level. One sees this at grad programs in math, where strong US departments may insist on stellar GRE scores -- but then proceed to chuck out 50-80% of their graduate intake because the students lack all sorts of other (untested) qualities. All that these "filters" are accomplishing is encouraging brutal, mindless competition in areas that at best constitute only one part of core competency.

The Europeans probably have a more sensible attitude about all this. Thus Oxford, for example, bases its admission primarily on an interview. I personally know of cases where they've made offers asking for only two "E" grades at GCE "A" levels -- something inconceivable at Harvard or Princeton (unless the applicant is a "legacy scholar" like George W Bush or John Kerry).

Just my opinions, as always.

 

[DominiConnor]

Actually, I've just noticed that the ad on this page talks of Baruch's 14% acceptance rate, so clearly MFE programmes know that exclusivity has value. You would not advertise a film as "Die Hard 17, only 14% of people allowed to watch it", though as I understand it, Basic Instinct 2 actually showed to empty cinemas in some places.

Wolf points out an issue that bugs us a little, the high drop out rates. We do not have great statistics, but we have a lot of anecdotes of 25-50% of students not completing. But of course they don't get their money back.

I agree that pure scores of any kind are far from perfect predictors of eventual ability, the difference between the Nth and the N+1th percentile is well within the noise of the process.

Also, although this is a math based subject, that is only part of the work, you guys are going to be spending 60% of your working life fighting computers. Does any MFE programme anywhere test for programming ability ? Can you articulate your new ideas ? Do you ever have new ideas ? Are they any good ? Can you interpret data ?

But you cannot hope to interview everyone, it would take weeks, and although Wolf approves of the British system, it is really struggling to cope with the vast increase in the number of applicants. 95% of all students are accepted simply on the basis of test scores, tests that are widely believed to be bogus.

I venture to say that any MFE programme that interviews all applicants will get far more people trying to get in. Unless programme want to hire professional full time interviewers, it ain't gonna happen. You simply can't get good FinMath teachers who will put up with interviewing 2,000 students, most of which can't do this stuff.

 

[bigbadwolf]

But you cannot hope to interview everyone, it would take weeks, and although Wolf approves of the British system, it is really struggling to cope with the vast increase in the number of applicants. 95% of all students are accepted simply on the basis of test scores, tests that are widely believed to be bogus.

I venture to say that any MFE programme that interviews all applicants will get far more people trying to get in. Unless programme want to hire professional full time interviewers, it ain't gonna happen. You simply can't get good FinMath teachers who will put up with interviewing 2,000 students, most of which can't do this stuff.

You're quite right: filters are in place by both HR departments and academic departments to reduce mountains of applications to more tractable hillocks (or simply to dissuade certain kinds of people from applying). But as you've rightly pointed out, such filters are not always accurate indicators of future performance, and may also leave out people who would otherwise be good bets. Perhaps necessary and inevitable evils that are a concomitant to automated procedures in our contemporary mass societies.

 

[Tigga]

But you cannot hope to interview everyone, it would take weeks, and although Wolf approves of the British system, it is really struggling to cope with the vast increase in the number of applicants. 95% of all students are accepted simply on the basis of test scores, tests that are widely believed to be bogus.

You are right. They can't not interview everyone.
Some only interview those they are interested in. They do phone interview for int'l applicants and in person for local applicants. Baruch, CMU, UCB do this.
Others just rely on test score and send decision without interviews. Columbia and a few other programs don't interview.
At least that's what I heard.

A new trend is to ask applicants to submit a video file where the student sits in front of a camera/webcam, and talks stuff for a few minutes. Burn that file to a cd and send in with application.
This saves quite a lot of money and time for the school. some even interview on Skypes
UCB and Michigan (not sure) do the webcam thingy.

In a few years, they probably will ask applicants to send in DVD resume or post on youtube showing their powerful tennis serve.
And we all have Aleksey Vayner to thank for.