Business schools in disarray

[bigbadwolf]

Interesting piece in today's FT on the disarray and crisis of confidence in business schools:

The economic crisis has thrown business schools into disarray. Deans and professors are suffering a crisis of confidence and are seriously questioning the structure and content of a twenty-first century management education. At the same time, business schools are searching for a way to contribute to global economic recovery. How can trust be rebuilt in business schools in general and the MBA programme in particular?

One would wish the same could be said of MFE schools/departments but that would be wishful thinking: the majority are run cynically as money-making endeavors with no critical introspection on the responsibilities and social worth of what they do or of the people they turn out.

Are financial engineers just glorified but soulless number-crunchers and programmers who obediently do the bidding of their masters? Yes, I suppose so. The disarray and crisis of confidence in business schools is partly because they are training business leaders who (theoretically) have economic and social responsibilities that transcend their company's bottom line. The MFE is a narrow technical specialist. I'm saying nothing new here. Yet at the same time one could wish for a bit more critical introspection on whether the risk-management tools taught work, to what extent they work, and when and where they break down. Indeed, do they work at all? Are they meant to work? Or are they just intellectual camouflage?

 

[IlyaKEightSix]

Well, for the nth time, I'll reiterate:

If one is going to bash financial engineering, and the use of advanced, higher-IQ techniques as applied to finance, EG programming, a zillion different statistical and operations research techniques, then one only needs to look towards the Medallion fund (RenTec), D.E. Shaw, perhaps Bridgewater, Edward O. Thorp, Emanuel Derman, et al. and realize that financial engineering is the future, and it will not be stopped.

Anyone who criticizes that quants are full of it (EG Taleb), despite all of their points for a majority of cases, should also realize that most FE schools are also garbage.

I know in Lehigh's case, their "MFE" program is trash. Steaming, stinking, rotting, fetid trash. They have a stochastic calculus course, two optimization courses, and a stochastic processes course. Case in point, I nearly completed all of that aside from stoch. calc because I wasn't able to sit in on it due to logistical reasons, and I recently applied to Lehigh's statistics masters, NOT its analytical finance masters.

That said, the professors do care about students that are willing to learn. I know that my favorite professor covers more credits than anyone else for that program, and she's an absolute angel and I'm hoping to be an industry contact for her in the near future.

As for the B-schools, the real question is this:

Jim Simons has no MBA.
David Shaw has no MBA.
Sergei Brin has no MBA.
Bill Gates doesn't even have a BS.
Warren Buffett has no MBA.
Emanuel Derman has no MBA.

However, at some point, all of the bozos at the top of today's bulge brackets DO have MBAs. So at some point, what the quants do is lost in translation.

I remember a person in my university (investments office guy) actually met Henry Laufer, and Jim Simons's #3 as well. After finding out that they had PhDs in things ranging from chemistry to astrophysics to economics and from the double digits in terms of different countries, he asked this question:

"How do you all communicate, then?"

The answer?

"Math."

The problem isn't with the financial engineers. The problem is with the fact that financial engineers thrive in an environment in which PhDs and MFEs rule. The mistake that the BBs make is thinking that quants and managers mix. They mix as well as water and electricity. It's an absolutely different way of thinking. Management wants transactions and money now, quants want to do research and go at their own pace.

In fact, Simons scoffs at the pay-per-quarter performance when he himself acknowledges that a research project can take *YEARS* to put together. Heck, it could take a long time just for an idea to be seen as wrong! (From what I hear, new hires troll through all of the files and see what works--and probably a lot of stuff which didn't!)

Quants like their think tanks. Managers and business guys like their hustle and bustle. Somewhere in a happy medium are the traders that like a bit of both.

 

[bluechimp]

Ilya, do you even know what you're talking about? Your overzealousness towards 'advanced, higher-IQ techniques as applied to finance.. statistical and operations research techniques' is really annoying. Last time I checked with my buddies at two of the funds you mentioned, they completely trash (literally) any sort of investment theses that is not backed by some real business/economic explanation behind it. Those funds also hire people based on how intelligent he/she is, not by the denomination of their degrees. I've seen very smart, successful and humble MBA-holders, as often as I've met nerdy wannabe financial engineers that think everything can be represented by equations and solved by linear programming. (Here's a trivia: do you even know what's the underlying applied mathematics field associated with Renaissance's quant strategy?).

Anyway, to respond to the original query (sorry I get sidetracked easily -- this happens a lot with inane comments):

Although I don't think there is an active push to change the curriculum within the space of MFE programs, I guess it's also unfair to say there will be zero response to the fact that some of the very popular quantitative models were used recklessly leading to the current financial crisis. I am not familiar with Baruch's program, but NYU and Berkeley have instructors that hail from the industry, too. I don't think they completely turned a blind eye to how things have turned out in their workplace, and consequently the lessons from the crisis will (hopefully) be passed to the students. It's hard to teach prudence in a classroom setting, especially with so much confusing materials to cover.. Also, I think higher now of programs like Princeton's, which has strong focus on economic modeling of financial systems and also interesting courses on policy-making, both vital elements in wake of this crisis.

Look, I think it's a more subtle question than just 'do the models work?'. I think to some extent, the VaR model used was a good start to approximate risks. But really, the best risk managers in the business knew that you cannot rely on mathematical models alone. I spoke once to Rick Bookstaber, and he totally ripped on people who depends completely on models. JPMorgan built a 'fortress balance-sheet' (a phrase Dimon really really loves) and fared better than rival banks, primarily because they thought exuberance in the markets did not make sense, not only because some red light was flashing on their cockpit. It is really hard to model panics and manias, you rely on your common sense and prudence in tandem with your models. Then you hope that you got most of your thinking correct, and that nothing blew up on your face.

To that end, though, I don't know how you can teach that kind of real-world experience via these professional programs.

 

[bigbadwolf]

Look, I think it's a more subtle question than just 'do the models work?'. I think to some extent, the VaR model used was a good start to approximate risks. But really, the best risk managers in the business knew that you cannot rely on mathematical models alone. I spoke once to Rick Bookstaber, and he totally ripped on people who depends completely on models. JPMorgan built a 'fortress balance-sheet' (a phrase Dimon really really loves) and fared better than rival banks, primarily because they thought exuberance in the markets did not make sense, not only because some red light was flashing on their cockpit. It is really hard to model panics and manias, you rely on your common sense and prudence in tandem with your models. Then you hope that you got most of your thinking correct, and that nothing blew up on your face.

Thank you for your nuanced and thoughtful answer, which has been a pleasure to read.

 

[Sanket Patel]

Jim Simons has no MBA.
David Shaw has no MBA.
Sergei Brin has no MBA.
Bill Gates doesn't even have a BS.
Warren Buffett has no MBA.
Emanuel Derman has no MBA.

You know who else doesn't have an MBA? LeBron James.

 

[IlyaKEightSix]

Ilya, do you even know what you're talking about? Your overzealousness towards 'advanced, higher-IQ techniques as applied to finance.. statistical and operations research techniques' is really annoying. Last time I checked with my buddies at two of the funds you mentioned, they completely trash (literally) any sort of investment theses that is not backed by some real business/economic explanation behind it. Those funds also hire people based on how intelligent he/she is, not by the denomination of their degrees. I've seen very smart, successful and humble MBA-holders, as often as I've met nerdy wannabe financial engineers that think everything can be represented by equations and solved by linear programming. (Here's a trivia: do you even know what's the underlying applied mathematics field associated with Renaissance's quant strategy?).

Do you mean stat-arb? Risk/reward/correlation/time series/statistics?

Because they do not let on their strategies.

 

[Stefan Zota]

Ilya, do you even know what you're talking about? Your overzealousness towards 'advanced, higher-IQ techniques as applied to finance.. statistical and operations research techniques' is really annoying. Last time I checked with my buddies at two of the funds you mentioned, they completely trash (literally) any sort of investment theses that is not backed by some real business/economic explanation behind it. Those funds also hire people based on how intelligent he/she is, not by the denomination of their degrees. I've seen very smart, successful and humble MBA-holders, as often as I've met nerdy wannabe financial engineers that think everything can be represented by equations and solved by linear programming. (Here's a trivia: do you even know what's the underlying applied mathematics field associated with Renaissance's quant strategy?).

Anyway, to respond to the original query (sorry I get sidetracked easily -- this happens a lot with inane comments):

Although I don't think there is an active push to change the curriculum within the space of MFE programs, I guess it's also unfair to say there will be zero response to the fact that some of the very popular quantitative models were used recklessly leading to the current financial crisis. I am not familiar with Baruch's program, but NYU and Berkeley have instructors that hail from the industry, too. I don't think they completely turned a blind eye to how things have turned out in their workplace, and consequently the lessons from the crisis will (hopefully) be passed to the students. It's hard to teach prudence in a classroom setting, especially with so much confusing materials to cover.. Also, I think higher now of programs like Princeton's, which has strong focus on economic modeling of financial systems and also interesting courses on policy-making, both vital elements in wake of this crisis.

Look, I think it's a more subtle question than just 'do the models work?'. I think to some extent, the VaR model used was a good start to approximate risks. But really, the best risk managers in the business knew that you cannot rely on mathematical models alone. I spoke once to Rick Bookstaber, and he totally ripped on people who depends completely on models. JPMorgan built a 'fortress balance-sheet' (a phrase Dimon really really loves) and fared better than rival banks, primarily because they thought exuberance in the markets did not make sense, not only because some red light was flashing on their cockpit. It is really hard to model panics and manias, you rely on your common sense and prudence in tandem with your models. Then you hope that you got most of your thinking correct, and that nothing blew up on your face.

To that end, though, I don't know how you can teach that kind of real-world experience via these professional programs.

Good answer!

In my opinion MFE programs have the potential to adapt quicker than MBA to changes in the environment. They start from a set of requirements (stochastic calculus, statistics, linear algebra etc). From this you can branch in different theoretical/practical directions.

A model is not being "preached" or "proven". It is presented with +/- including calibration, implementation etc. No decision is taken, in fact the quants have a position of "data providers" in many shops. Somebody else decides between different results.
In this case, the "implications", "ethics" and others are less relevant.

 

[bigbadwolf]

In my opinion MFE programs have the potential to adapt quicker than MBA to changes in the environment. They start from a set of requirements (stochastic calculus, statistics, linear algebra etc). From this you can branch in different theoretical/practical directions.

I think the situation now (and perhaps also in the past) is akin to what has been happening in theoretical physicists: the most important work is being done by people who are systematically and carefully working through the foundations, looking for new tools, groping for new insights, and having a philosophical mind-set. These people may or may not be math whiz kids -- but even if they are, it's just one more string to their bow. I'm distrustful of whiz kids who come armed with technical virtuosity but little else.

 

[bluechimp]

Do you mean stat-arb? Risk/reward/correlation/time series/statistics?

Because they do not let on their strategies.

No, no, no. And this was on the WSJ yesterday (please, tell me that you read the WSJ regularly? not doing so is a recipe for destruction, to say the least). Do the acronyms HMM or NLP ring any bells? Now I understand why do you deify Jim Simons: just like God, Jim Simons is incomprehensible in why or how He does things! (disclaimer: the writer of this post is comfortably ambivalent about his religious beliefs).

My point is, do not go out there and shout out stuff that you apparently know very little about. Please, it is indicative of ignorance, laziness, and lack of depth in what you have learned. Also, allow me to point out something about Renaissance, by quoting WSJ (the same article):

For much of its 27-year history, Renaissance has been among the most envied hedge-fund firms in the world, as Mr. Simons's investing success has made him a billionaire several times over and one of the wealthiest money managers anywhere.

But Renaissance clients have pulled billions of dollars since early 2008 as performance has suffered in its biggest funds open to outside investors. Firmwide, assets have shrunk to about $17 billion from a mid-2007 peak of some $36 billion. A person familiar with the firm said requests by investors to pull money have slowed lately.

Meanwhile, Mr. Simons and fellow Renaissance executives have faced questions from some clients over the disparity between their returns and the consistent big profits of Medallion, a fund started in 1988 that is now held primarily by Renaissance insiders. Renaissance executives say Medallion employs a different strategy than the newer funds.

I hope it's not another Maddoff scheme in the making.

 

[Sanket Patel]

My point is, do not go out there and shout out stuff that you apparently know very little about. Please, it is indicative of ignorance, laziness, and lack of depth in what you have learned..

Well said, sir. Well said.

 

[bigbadwolf]

My point is, do not go out there and shout out stuff that you apparently know very little about. Please, it is indicative of ignorance, laziness, and lack of depth in what you have learned.

Relax. You are not going to change him. The trick is to let your eye elide over his posts. In your life you have doubtless met many individuals who talk a great deal without communicating anything of significance and you have must have learnt to "switch off." The same thing here.

For those capable of learning, a gentle hint is enough. Otherwise, all the lecturing in the world will fall on deaf and unresponsive ears.

 

[IlyaKEightSix]

No, no, no. And this was on the WSJ yesterday (please, tell me that you read the WSJ regularly? not doing so is a recipe for destruction, to say the least). Do the acronyms HMM or NLP ring any bells? Now I understand why do you deify Jim Simons: just like God, Jim Simons is incomprehensible in why or how He does things! (disclaimer: the writer of this post is comfortably ambivalent about his religious beliefs).

Hidden Markov Models and Non-Linear Programming, thanks. My major was an OR one. I'm not sure I was ever exposed to HMMs, and NLP isn't actually too bad. Except here's the thing--a lot of quant shops use similar underlying mathematics. But RenTec has done spectacularly with Medallion.

As for "another Madoff scheme", there are hundreds of hedge funds with X different strategies. Plus, since Medallion is the employee fund, that explains the reason that it has the quality it does. You'd think that employees would care about their own holdings most.

 

[Aditya Chitral]

Relax. You are not going to change him. The trick is to let your eye elide over his posts. In your life you have doubtless met many individuals who talk a great deal without communicating anything of significance and you have must have learnt to "switch off." The same thing here.

For those capable of learning, a gentle hint is enough. Otherwise, all the lecturing in the world will fall on deaf and unresponsive ears.

Atleast the questions make a person aware of what he knows and what he does not. That is a sign of a true student. While the people who prefer not to shout always remain ignorant. The day you start assuming you know everything and the person next to you knows nothing, you are creating a "PACKAGE OF DESTRUCTION".

And I am sure a person would not be overly disturbed and worried if does not have testimonials from some chimps and wolves on his resume.

 

[bluechimp]

Hidden Markov Models and Non-Linear Programming, thanks. My major was an OR one. I'm not sure I was ever exposed to HMMs, and NLP isn't actually too bad. Except here's the thing--a lot of quant shops use similar underlying mathematics. But RenTec has done spectacularly with Medallion.

As for "another Madoff scheme", there are hundreds of hedge funds with X different strategies. Plus, since Medallion is the employee fund, that explains the reason that it has the quality it does. You'd think that employees would care about their own holdings most.

Yes, they do use HMMs extensively. But unfortunately, NLP does not stand for non-linear programming. Did you ever try to dig up the article I quoted? The answer is right there, in front of your eyes. Come on now, I'm making it easy for you. Oh, and you say that a lot of quant shops use the similar underlying maths? What are these principles? How do you use them to make investment decisions? Please, use your superior communication skills and teach us all about how they work!

Relax. You are not going to change him. The trick is to let your eye elide over his posts. In your life you have doubtless met many individuals who talk a great deal without communicating anything of significance and you have must have learnt to "switch off." The same thing here.

For those capable of learning, a gentle hint is enough. Otherwise, all the lecturing in the world will fall on deaf and unresponsive ears.

I am sorry if I go into seemingly unnecessary ad hominem attacks. I'll be the first to acknowledge that I have a very short fuse against ignorant people; it's pretty bad since, as you all can see, I get riled up just because a post made by some anonymous nobody on the internet. This peculiar attribute of mine has been both a blessing and a curse, really.

 

[Sanket Patel]

As Plato said, "wise people talk because they have something to say; fools, because they have to say something."

 

[IlyaKEightSix]

Hidden Markov Models and Neuro Linguistic Programming. In other words, while RenTec is still doing stat-arb, it's honestly trying to quantify human behavior by interpolating it through security prices? Anyhow, yes, I did read that whole article. However, I have never heard of HMMs or NeuroLinguistic Programming (at least by those names) prior to today. And I never said that I knew everything. I know extremely little and have a long way to go before I can do what I dream of doing, and I realize that.

So thank you for that enlightening note. Interestingly enough, I have been exposed to basic neuro-linguistic programming (didn't know it by that name) in a game design class when talking about finite state machines in video games.

As for MBAs...I'm sure there are some very hard working and successful MBAs. But asking an MBA to know what a quant is doing is like asking a quant to know the ins and outs of the softer side of finance--the boardroom dealing, the rainmaking, etc... Now I'm sure there are those who are so exceptional that they can do both.

 

[bluechimp]

Hidden Markov Models and Neuro Linguistic Programming. In other words, while RenTec is still doing stat-arb, it's honestly trying to quantify human behavior by interpolating it through security prices? Anyhow, yes, I did read that whole article. However, I have never heard of HMMs or NeuroLinguistic Programming (at least by those names) prior to today. And I never said that I knew everything. I know extremely little and have a long way to go before I can do what I dream of doing, and I realize that.

So thank you for that enlightening note. Interestingly enough, I have been exposed to basic neuro-linguistic programming (didn't know it by that name) in a game design class when talking about finite state machines in video games.

As for MBAs...I'm sure there are some very hard working and successful MBAs. But asking an MBA to know what a quant is doing is like asking a quant to know the ins and outs of the softer side of finance--the boardroom dealing, the rainmaking, etc... Now I'm sure there are those who are so exceptional that they can do both.

You are so lost. The fact that Renaissance hired a lot of speech recognition specialists should've given it away. The definition of neuro-linguistic programming on wikipedia is so ridiculous that I won't even bother to paste it here. In poker parlance, you don't scare off a big fish by knocking on the fishbowl. That is, you don't tell a poor player how bad he is performing so that the everyone can wipe him off the table. Maybe I should just stop responding to this thread from this point on.

At any rate, Ilya, sorry if I came off very harshly in my previous posts. Look, as a student, I've always struggled with large gaps between how much I know, how much I think I know (more inflated than my actual knowledge), and how much I need to know to perform well on the job (absolutely much bigger than the previous two variables). I try to make sure that those gaps are getting smaller by the day; and to that end, I never felt I completely understood anything. In contrast, I felt that your posts did not resonate with your claims that you 'know extremely little' -- although if I made mistakes in interpreting them, then mea culpa.

Now that I think I have gotten my points across to you via grilling, I will drop the issue of hounding you down with feeble matters to give way to more constructive discussion. For all my incendiary comments, QN members, that was most definitely mea maxima culpa.

 

[Liquid Assets]

The softer side of finance is what a quant should know. A quant is not just a person that's good at math or computer programming. As quants we put ourselves down and limited our abilities to be more in the work place. Example. I work in risk management and at times I sit at the table with salemen/women while they are talking to clients. I' m at the table not only as the person underwriting the deal or calculating the probability of default, but also to assist in closing the deal. Why? Because I can articulate my thoughts/finding with the numbers, but also I am able to communicate the pros and cons of this and that. Plus, I am able to effectively write why this deal makes sense verse why it does not. These are very very very important skills to master. Finance is an art as well as a science. Art, communication of deals, write-up, persasives speechs on why our firm is better than ABC or XYZ, understanding the client and their needs etc. Science, numbers, understand the numbers and why the out is what it is what are the inputs etc.