CMU MSCF students at the 55 Broad Street campus

Carnegie Mellon University (CMU) announces today that it will offer an online version of its Master in Computational Finance (MSCF) program, to be called MSCF ONLINE starting in Fall 2012.

Among the first universities to offer a master degree in financial engineering (MFE) degree in the US during the 90s, CMU MSCF program attracts a large pool of applicants vying for 90-plus slots available at both main campus in Pittsburgh and its downtown NYC campus. For the 2011 admission year, CMU MSCF received an all-time high 1,123 applications, most among all MFE programs.

While not the first university to do so, the decision by CMU to offer a quantitative finance online degree marks a new phase in the increasingly competitive MFE education landscape. University of Southern California recently introduced an online MFE degree, delivered through Viterbi School’s Distance Education Network (DEN). Stevens University is another offering an online certificate and degree equivalence of its Financial Engineering program. University of Washington offers two online certificates, one in Computational Finance and another in Financial Risk Management.

These are not the only educational options available online to anyone unable to take the usual night-time courses for part-time students. Certificate in Quantitative Finance (CQF), a $20,000 six-month online program run by 7City and started by Paul Wilmott, claims to “educates many hundreds of people every year, all around the world, in the highest level quantitative finance“. Despite the fact that CQF does not lead to a Master degree as those offered by many aforementioned universities, the low cost, short-time, access to online material, contribute to attract many to enroll.

Rick Bryant, Executive Director of the CMU MSCF program sounded plenty optimistic in his conversation with Quantnet. “We believe high quality, on-line instruction will continue to gain traction and are persuaded that this option will meet the needs of many individuals keen to develop their knowledge and skills in the field of quantitative finance but cannot relocate to Pittsburgh or New York”, he said.

Long known for its use of live telecast lectures between Pittsburgh and NYC campus, CMU MSCF plans to leverage the technology to give MSCF ONLINE students the most bang for their buck. Besides giving online students access to live and recorded video lectures, the program is actively exploring other modern technological venues such as forum for homework discussion, social media for networking.

With an estimated $3,228 per course, the 25-course MSCF ONLINE would cost close to $82,000, the most expensive among similar online offerings. Adding to the overall cost is a list of required hardware that all ONLINE students must obtain. Among them are a projector, a scanner, a HD webcam, and an approved computer. And no applicant who lives within 50 miles of either Pittsburgh or NYC campus need apply.

Realizing the obstacles to attract students to the MSCF ONLINE, CMU plans to pilot test the program on few applicants. Qualified applicants with a complete application by August 22 may be able to attend this Fall. This would allow CMU MSCF to work out technological kinks and logistical problems before Fall 2012.

CMU certainly would not be the last university to venture into online quant education. In September, Quantnet will introduce its own online C++ certification for MFE applicants in partnership with the Baruch MFE program and author/trainer Daniel Duffy.

With much of the world’s communication, entertainment, and learning gradually moves online, it’s not hard to imagine a future world where one would get an entire post-baccalaureate education online, even in such a highly specialized as financial engineering. Whenever it will be, it will certainly be just as competitive as their traditional counterparts.

Baruch MFE classmates, myself in the Quantlab

It’s been a while since my last post.  It has been quite the semester. I am currently visiting my parents and girlfriend in Canada during our winter holidays. I figured I would take the time to talk about my first semester in the Baruch Financial Engineering program.

The first semester included taking 4 graduate level courses. These were Stochastic Processes in Finance, Numerical methods in Finance, Object Oriented programming in Finance and Pricing of Financial Instruments. All the courses were heavy on the quantitative side. The Pricing course included a strong exposure to lots of financial products which was a good relief from the heavy math in the other courses. Every course included lots of programming (C++ especially) except Stochastic Processes class.

The most challenging class definitely had to be Stochastic Processes. There is a big focus on Measure Theory and Real Analysis in this course. The second part of this course will be next semester which is supposed to include a significant amount of Stochastic Calculus and finance exposure. After going through this course, martingales, sigma-algebra, LDCT, MCT and Brownian motion have been embedded into my brains pretty deeply.

One of my favorite classes had to be the Pricing class. We had an amazing professor who had a lot of deep industry knowledge about various products. I definitely got exposure to a huge range of financial products in this course and how to price them. I can now effectively price several exotic options, vanilla options, FRA, FRN, CDS, Interest Rate Swaps, FX Swaps, etc. The course also gave me a lot of exposure to various risk management parameters like Delta, Gamma, Vega, Theta, etc. The course included a lot of coding in Excel/VBA.

The highest amount of programming in the course was done in two courses: Numerical Methods in Finance and Object Oriented Programming. Numerical Methods mostly included lots of Monte Carlo simulation and Finite Difference methods to price options. The course went through several finite difference methods which included Forward and Backward Euler, Crank Nicholson and many more. We priced vanilla options with dividends, without dividends, with discrete dividends, with continuous dividends, exotic options, and many more products. This course included a significant amount of algebra, differential equations and programming. All the programming was done in C++.  Our final basically included pricing a Barrier option and an American option.

The last, but not the least, was Object Oriented programming in finance. This course included teaching us the intricacies of C++ at a much technical level. It included us learning about classes, pointers, polymorphism, STL, factories, etc and implementing various pricing procedures using optimized code. The course was taught by a programmer who works on the prop desk at Bank of America.

Well enough about courses. I have been applying for summer internships and full-time positions extensively. I also have been networking a lot by going to several conferences in the area. The fruits of networking and applying have already been paying off. I recently had the pleasure of interviewing with a large British bank for an internship, a private equity firm for a full time position, 2 hedge funds for an internship, a large Japanese bank for a full time position, a full time position at a large American BB, a full time position at a prop firm and a few other positions. I have a few offers now from the interviews. I cannot give out any more details for obvious reasons right now. Once it is all confirmed and set I will be sure to let you guys know.

For job seekers out there I will give you some information in how I secured these interviews. Mostly what I did was have an effective resume and just applied on websites and job boards. I also sent out several emails that I have collected over the last few months. These emails are mostly of people I met at conferences and events around New York. In addition, I contacted people who were part of the fraternity I am a member off from Wisconsin that are in finance, especially in New York. I think one of the biggest help in me getting interviews was probably the Resume Book that my professor sent to firms for internship positions. Other than that, I am sure some luck helped me too. I just stayed positive and highly pro-active and made sure I applied everywhere I could think off.

These two articles helped me quite a bit:

• How to get quant jobs -advice from Wall street executives
• List of firms that offer quant internships and jobs

I think I have a decent enough resume to get to the first round. Beyond that I believe the strong first semester at Baruch and general inter-personal skills aided me a lot in converting interviews to offers.

The social life at Baruch is absolutely great. My classmates and professors usually get together for drinks once a week. We even went to several famous spots on school field trips in New York which included the Metropolitan Opera, Museum of Natural History, Central park, etc.

I feel very blessed past few months. I have had the chance to make some great friends in the program, got to meet some great people in the industry, and learn a lot. I am looking forward to next semester. I will be taking Stochastic Calculus 2, Numerical Methods 2, Risk Management, and am still debating between Market Microstructure and Structured Finance. I would love to learn more about the role that exchanges play in trading. At the same time being an engineer, I love the notion of “creating” new products. Right now Structured Finance seems absolutely dead, but I do see a strong come-back in the near future. It is a hard decision. Let me know if you guys have any tips?

Until next time…

What are the best-selling books that Quantnet members purchased in 2010?

The following 25 best-selling books is compiled from a list of almost 600 titles purchased as provided by Amazon where Quantnet.com is shown as a referrer. The complete list is attached here.

By STEVE SHREVE

In his article “The flawed math of financial models”, Financial Times, November 29, Pablo Triana seeks to fix a large portion of blame for the world-wide financial crisis on “quants” in the finance industry and the programs that educate them. Mr. Pablo recommends radical reform in such programs. Others, carrying these ideas farther, call for a diminished role for quants in finance.

Any discussion of quants in finance must begin with the recognition that the global integration of economies and the associated complexity of our financial system has made the use of mathematical models an indispensable tool. Rules-of-thumb and intuition will not suffice when multi-national firms face exchange rate risk, funding risk and commodity price risk, when insurance companies and pension funds face longevity risk, when financial institutions are called upon to mediate these risks, and when regulators are charged to oversee these institutions. This was recognized in the recent U.S. financial reform legislation, which authorized a government Office of Financial Research whose task in 2008 would have been to alert policy makers to the ridiculously large naked position in credit default swaps held by AIG and to predict the effect of the failure of Lehman Brothers. Such an office must necessarily be populated by quants, people who can build models into which information about financial institutions is fed.

What then is the appropriate training for quants? I believe we should focus on three aspects.

Most importantly, a quant must be competent in the technical disciplines of mathematics, statistics and computer programming, and she must be knowledgeable about financial markets. Achieving competence across this broad spectrum is a tall order. But it must be done because a well-intentioned incompetent quant is as dangerous to the financial system as a well-intentioned incompetent doctor is to personal health. The primary focus of the educational programs at Carnegie Mellon will remain the creation of competent graduates. This is what we do best.

But a good quant also needs good judgment. A wise quant takes to heart Albert Einstein’s words, “As far as the laws of mathematics refer to reality, they are not certain; and as far as they are certain, they do not refer to reality.” All models are wrong. Judgment is needed to know when an admittedly wrong model can be helpful and when it is dangerous. This kind of judgment is acquired primarily through experience, but we can begin teaching it in the classroom. Since the financial crisis, we have invited participants in the crisis to speak in detail to our students about deals that went bad, describing how the deal was analyzed, why it was approved, and what was overlooked.

Finally, we need people with integrity managing our financial systems. Teaching ethics is difficult, and guaranteeing that listeners will implement those teachings is impossible. It is not easy for a quant to sound the alarm that his models are being stretched beyond their limits, knowing that if he is taken seriously it will result in the loss of business to competing firms and may result in the loss of his job. We cannot instill in sixteen short months behavior that properly requires years of nurturing and mentoring. We do what we can, leading by example, penalizing students for academic dishonesty, setting and enforcing rules for ethical conduct when interacting with potential employers, posing ethical dilemmas for classroom discussion, and encouraging our graduates to consult with fellow graduates when facing tough ethical decisions.

A lesson that can be learned from the present crisis is that if everyone implements the same good idea, their collective action can invalidate the assumptions that made the idea good. If everyone assumes that U.S. housing prices cannot decline and makes large bets based on that assumption, their collective action will ultimately bring about a decline in housing prices. This is not a new lesson; it is the lesson of every bubble. A feature of the most recent bubble is that quantitative analysis contributed to a false sense of security that encouraged firms to scale up risks. In some cases senior managers and even quants themselves did not appreciate the limitations in the models on which they based their risk analysis. Our students do not begin their careers at the level where the disastrous decisions were taken, and only a handful of them will ever reach those positions of power. Nonetheless, in the short time they are in our care, we seek to the extent possible to make them competent quants who exercise sound ethical judgment.

About the Author

Steve Shreve’s books The Binomial Asset Pricing Model and Continuous-Time Models are the required textbooks for many MFE programs’ Stochastic Calculus courses. He is a professor at Carnegie Mellon University and one of the co-founders of the M.S. in Computational Finance at Carnegie Mellon.

The following is a response by Mr. Triana sent to Quantnet on Jan 10, 2010.

By PABLO TRIANA

Let me first say that I deeply admire Professor Shreve. Though my mathematical background does not empower me to fully appreciate his scientific prowess (not that his unparalleled global reputation would ever necessitate my feedback as further support), I am aware that in replying to his analysis of my recent FT article on quant education I am addressing most possibly the world´s leading light when it comes to stochastic calculus and mathematical finance. And far from an aloof researcher, Professor Shreve is also a very successful and ingenius academic entrepreneur, having taken a leading role in the development and management of one of the most exciting and path-breaking university graduate programs ever devised. To top it all, I can personally attest to his human generosity and kindness, getting misty-eyed as I recall the time when Professor Shreve, now about a decade ago, kindly accepted my invitation (as President of NYU Stern´s Financial Engineering Association) to regale us with a wonderful lecture and an even more pleasant follow-up dinner at a fancy Soho restaurant. I vividly recall him being impressed by my thorough knowledge of and interest in his legendary Computational Finance program at Carnegie Mellon, to the point of asking me why I had chosen NYU instead (I didn´t even try to apply to terrifyingly intimidating Carnegie Mellon, acutely aware of my negligible chances at getting in; I ain´t no rocket scientist, folks!).

In sum, it is not only my responsibility but also my pleasure to try to address Professor Shreve´s rebuttal as respectfully as possible, given the caliber of the counterparty. I hope I manage to succeed at this, if not so much at triumphing in the debate.
Some initial clarifications are in order. I don´t really blame quants and quant programs for the crisis. I blame the use of certain models for the crisis. I don´t really care if those using, peddling, and imposing the deleterious models were quants, traders, salesmen, or fast food caterers. My goal is not to target specific groups of people, my goal is to target specific analytical concoctions. Having said that, it is true that a lot of quants vouch for those models both inside and outside the financial industry and, much more critically, vouch fanatically for the quantification of finance in general. As long as such belief is held and enthusiastically pushed, we can get in trouble because the potential for bad models to infiltrate the markets would be made that much larger. We need to create much more restrictive filters when it comes to welcoming mathematical finance wizardry into the realm of practice. Quants and quant programs could and should have been much less permissive and much more critical. Roadblocks to dangerous models should have been forcefully built by those who best understand the mechanics. So, yes, quants and quant programs could in the end be subjected to one type of accusation: neglect.

Everything stated in Professor Shreve´s response makes a ton of sense, and one can´t help but wholeheartedly agree. But, like other famed quants too graced with the ability to muse gracefully and the valour to challenge flawed quanty practices, Professor Shreve does not go far enough. Just like Emanuel Derman, Paul Wilmot, or Ricardo Rebonato, Professor Shreve needs to get closer to Nassim Taleb (and, maybe, my very humble self) and take things a step or two further and engage in a healthy dose of loud name-calling and unabashed denunciation. It is not enough to state that quantitative analysis played a role in the crisis by encouraging misplaced confidence or that many misunderstood the maths. It is imperative to endlessly fingerpoint the main culprits, essentially VaR and Gaussian Copula (to Professor Shreve´s credit, he went after the latter in a recent piece), and to make sure that such utterly failed tools are never again given the keys to the risk kingdom. Demonstrably flawed and lethal models should be banned from the land, and the real reasons for their original embracement intrusively inquired. VaR can no longer be part of banking regulations. These things can´t continue being taught, unless they are presented as the bad that can emerge from the quant lab. More pressing still, those failures must serve as catalyst to force everyone to revisit whether finance can and should indeed by mathematized. Are VaR, Gaussian Copula, Black-Scholes, Portfolio Theory, or Financial Econometrics isolated cases of failure, or rather inescapable proof that financial theory is bound to be at best useless and at worst crisis-igniting? We urgently need a Mathematical Finance Council of Nicaea, so that these pressing questions are answered once and for all. I wrote my Lecturing Birds On Flying in a naively idealistic attempt to help kick-start such process. Will the best that the discipline has to offer, like Professor Shreve, pick up the gauntlet?

This is no time for mincing words, it´s time to act. Back in 1994, Carnegie Mellon showed untold innovativeness and courage by correctly embracing the forcefully emerging field of financial engineering. It became the indisputable world beater. Now, with the discipline in tatters and accused of horrible crimes, the same institution should once more display one-of-a-kindness and lead the second quant finance revolution, the one that ought to make sure that models and financial stability can coexist side by side and the one not afraid to terminally castigate those naughty analytical concoctions that wreak havoc.

About the Author

Pablo Triana is the author of Lecturing Birds On Flying: Can Mathematical Theories Destroy The Financial Markets? (Wiley)

Hill Center, Rutgers University

When Marc Fields, sales executive at a financial services firm, got accepted into one of the two master programs in quantitative finance at Rutgers University earlier this year, he was still unsure of the differences between them: “After being accepted (into the MQF program) I discovered the MSMF program at the New Brunswick location through the Math department. I was a little shocked at this discovery especially after looking extensively at Rutgers. Upon further investigation I still couldn’t make out the difference between them and which one would have been the right fit.”

A quick look at similar questions on Quantnet’s discussion board shows Mr. Fields is hardly alone. Next year, aspiring applicants like him will have a harder time making a decision between available options at Rutgers.

On Dec 16^th, 2010 Rutgers University Graduate Faculty Council gave the go-ahead to the Department of Statistics and Biostatistics’ new graduate program, titled “Financial Statistics and Risk Management” (FSRM). The program will begin admitting students in the Fall of 2011. The department has been working for the past two years to create this new program (website).

Response to news of the launching FSRM has been ambivalent, to put it mildly. At least two directors of quant master programs in the New York area were in disbelief when apprised of the fledgling FSRM program. Paul Feehan, director of Rutgers’ Master of Mathematical Finance (MSMF), released an official statement to Quantnet on behalf of his program strongly objecting to the launch of the new program: “The establishment of a third program, in what is generally viewed as a single field at the master’s degree level, is not in the best interests of the university or its students.”

“If there were three options within Rutgers for quant programs I would have been very confused. It is hard enough to define financial engineering programs between different schools, but to have three programs within one school seems to add another layer of complexity to the problem.” – said Mr. Fields when told of the new program.

The FSRM and MSFM (website, review) programs are both housed in Hill Center, Busch Campus – in New Brunswick/Piscataway.

FSRM is Rutgers’ third entry into an increasingly saturated market. The Master of Quantitative Finance (MQF), the other quant master program out of Rutgers’ Business School, is based in Newark, NJ.

When contacted by Quantnet, Rong Chen, Director of the FSRM program, told us that his program was created “to address the increasingly strong demand for specific training in statistical and data-oriented techniques that are widely used and help to drive success in the financial industry.” And there is some plausibility to this argument: in a Quantnet interview with several Wall Street managing directors, one executive at a top investment bank commented that MFE graduates tend to be strong in programming but relatively weak in statistics. This executive has been known to push for more statistics courses in several quant master programs where he has been teaching as an industry practitioner.

“We believe that there is a strong demand for well-trained master’s level employees with a solid methodological foundation and practical skill set in financial statistics and risk management. Indeed, especially in this economy, we believe that these skills will give our graduates a considerable edge in career opportunities. We expect to place our students well.” – said Mr. Chen. His program is expected to enroll 20-40 students in the first few years and more as it matures. All core courses will be offered in the evenings to accommodate part-time students. On its website, the program is posting a job position for an Associate Director of Administration who will manage daily operations as well as job placement for students.

How would applicants like Mr. Fields decide which to study when faced with choice of MSMF, MQF or FSRM?

“Obviously, I would have been unsure of the differences of each program and which program would better suite me for my career objectives. Aside from the differences in course load it would be good to know how the faculty and recruiting differ among the different options. It would also be great to know how the street views each program but that is a little harder to quantify.”- said Mr. Fields.

“I don’t want to assume that Rutgers or any school having three quant programs it necessarily bad but the school should make it abundantly clear what the differences between the programs are and market them as so. I think it’s a little ambiguous right now especially when there seem to be 10 new programs popping up everyday around the country. If you are creating a new program you have to explain how you will get your graduates jobs and how you differentiate and compete with similar programs.

Mr. Chen has a quick answer: “The graduates from the FSRM program will be able to take immediate employment as financial statisticians and risk managers. Students who are interested in mathematical modeling and stochastic differential equations should consider the MSMF program, and students who are interested in becoming financial managers should consider the MQF program. The FSRM program is most suitable for students who are interested in statistical methods, analyzing large financial data set, identifying statistical arbitrage opportunities, proprietary trading strategies, stress testing, and risk management.”

Even so, this industry desire for people with more statistics under their belt may not necessarily translate to a successful launch for Mr. Chen’s FSRM program. As Paul Feehan makes clear, no other university offers three similar programs: other universities tend to concentrate their limited resources into strengthening the one program choice they do offer.”

As if on cue, Columbia University’s Business School recently announced that it will offer a master’s degree in Financial Economics. Its website promises students “analytical jobs with potential employers such as investment and commercial banks, asset management companies, consulting firms, and policy-oriented organizations “.

“The existence of three quantitative finance master’s programs at Rutgers would confuse applicants, employers, and external program reviewers responsible for rankings”, Mr. Feehan of Rutgers’ MSMF said in the program’s response. When it comes to ranking and confused external reviewers, he has first hand experience. When AdvancedTrading, an industry magazine, released its “Top 10 Quant Schools” ranking in 2008, it listed both Rutgers MSMF and MQF programs. But according to our conversation with one person involved in the ranking, only one of the Rutgers program was submitted to the magazine. When AdvancedTrading mistakenly put the other Rutgers program in its ranking, instead of acknowledging and correcting the mistake, it decided to put both Rutgers programs on the list. As a result, there are 11 programs in its “Top 10″ list.

Aside from the confusion for applicants, universities offering competing programs inevitably face the perception that they are trying to cash in on the applicants, many of whom are international students who pay $50,000 or more for a master’s program. This perception may not be of vital significance to cash-strapped Columbia University, for example, which offers a Master of Financial Engineering (MFE) in the IEOR dept as well as an MA in Mathematics of Finance program, jointly offered by the Dept of Mathematics and Dept of Statistics: the MAFM program, perceived as inferior to Columbia MFE program, still managed to draw over 800 applicants for its Fall 2010 incoming class. Rutgers, however, does not have the Ivy League cachet or the New York City location advantage to attract domestic applicants, an increasingly prized group courted by programs new and old.

That issue was clearly on Mr. Fields’s mind has when he spoke candidly to Quantnet.

“My biggest concern with any program is the recruiting aspect. As more programs are created it makes it harder for recruiters to differentiate candidates and the programs they attended. Once this happens they will start to resort to looking at the school name as their basis for quality rather than the intricacies of the course load. Rutgers was able to get a top ten ranking and give it some validity which is great, but they should have taken that and run with it. Rather than increase class sizes and developing alternate programs they should have spent their resources developing their current programs and making sure that they continued to get mentioned and compete with other top 10 programs.”

“I would be much more comfortable with Rutgers granting one, maybe two, financial engineering degrees but having different concentrations. That way students who are math, statistics, or finance focused can still tailor their course load accordingly but not dilute the Rutgers reputation or make Rutgers split up resources.”

The competition that the FSRM will encounter from well-established programs will doubtless be fierce in the years ahead. The tri-state area is home to many of the oldest and largest quant master programs. Established programs like Carnegie Mellon’s Computational Finance (CMU MSCF) and Cornell University have set up remote campuses in NYC at 55 Broad Street. Baruch College, Columbia University, and NYU’s Courant are three other well-known programs in the city offering competing programs.

Mr. Yangru Wu, Director of the Rutgers MQF program (website, review), told Quantnet that his program does not share resources or collaborate with the new FSRM program at the moment. Mr. Chen sees it differently: “FSRM is closely collaborating with QFM program at Rutgers in sharing resources, including courses, financial data and joint activities.”

Besides sharing the same building, students in the MSFM and FSRM programs will likely see more of each others in the near future, Mr. Chen predicts: “We are also working with MSMF program to provide better learning experiences for the students in both programs. For example, for the three statistics courses that MSMF students are required to take, we will create special sessions exclusively offered to FSRM and MSMF students with extensive training in finaicial applications and enhanced instruction and learning support.

“The statistics department has been teaching courses as a part of the MSMF program since its existence and will continue to do so. We look forward to developing deeper connections with MSMF and MQF in the future, in order to further take advantage of common resources and build a broader community for quantitative finance at Rutgers” – said Mr. Chen.

Tuition for FSRM has not been finalized but should be similar to those of MSMF program, according to the program’s website. Current tuition rate for MSFM program is $839/credit (residents) and $1,314/credit (non-residents). At 30 credits, total tuition would be around $25K for residents and $39K for non-residents. For comparison, tuition rate for the Rutgers MQF is $877/credit (residents) and $1,446/credit (non-residents). Tuition for the 48-credit MQF degree will be $42K (residents) and $69K (non-residents).

At Baruch College, part of the public City University of New York, tuition for residents is only $11K and $22K for non-residents. At private universities in the area, a MSMF degree from NYU Courant would cost $52K while at Columbia University, students pay $52K for the year-long MFE degree or $39K for the 2-semester MAFN program.

A New Jersey native, Mr. Fields recently moved to New York City and decided to pursuit an MBA from Stern instead of the MQF degree. Explaining the change of heart, he said: “At the end of the day I thought that an MBA and CFA designation would be a better fit for me. My career goals have shifted over the past couple years and I think to be competitive, my time and resources would be better spent getting an MBA and CFA. Additionally I believe the Stern reputation and recruiting is an advantage over either of the Rutgers programs”

Speaking of his brief time in the program, he said: “I had only studied at Rutgers MQF program for one full semester and a summer session before switching to Stern, so my experiences should only be taken in context of my exposure to the program. I really enjoyed the Rutgers MQF program and thought that it was heading in the right direction. There are many qualified students studying there and the faculty and staff seem dedicated to providing a valuable education.”

This is the second year the workshop is offered by the Mathematics in Finance Workshop and Conference Center at Courant. According to an email sent to Quantnet in November, the workshop’s intended audience is “buy-side practitioners (portfolio managers and risk managers), sell-side practitioners (traders, financial engineers, quantitative analysts, research teams), and academics will deepen and broaden their understanding of the recipes they implement everyday and will learn the most cutting-edge techniques.”

The event’s materials list the following topics:

• Financial market microstructure for the practitioner and the mechanics of trading
• How to work with high frequency data
• Common trading strategies
• Estimation of transaction costs and market impact models
• Portfolio construction with the Black-Litterman model and robust optimization
• Portfolio optimization with transaction cost
• Simulation techniques
• Back-testing strategies
• Multi-period dynamic portfolio optimization with transaction costs

Lee Maclin, an adjunct professor at Courant Institute and a founding partner of Pragma Financial Systems, presented in detail the mechanics of trading and many of its available tools such as limit and market orders, the two-way double auction, direct market access pipes, etc. He described how they work, as well as the effect they have on the market and how it could influence our positions and decisions. He continued with ways of working with high frequency data and what types of data set can be constructed and used.

Petter Kolm, clinical associate professor at Courant, had the chance to introduce to the audience the calculus of variations and a couple of market impact models. He discussed the Almgren-Chriss model and the capabilities it has along with the the efficient frontier of optimal execution and models that could utilize public data.

The rest of the day included Petter Kolm and Farshid Asl, an adjunct professor at Courant Institute. Dr Kolm gave an interesting presentation of the Black-Litterman model and discussed robust portfolio optimization techniques. Farshid Asl, was up next giving a basic overview of financial time series modeling. He also explained filtering in statistical arbitrage and applications of Hidden Markov Models.

Courant workshop attendants discuss material with lecturer during break

The second day of the conference started with Eran Fishler, director of algorithmic trading at Pragma Securities, giving an overview of dark pools and how to trade using dark pools. He went on to include empirical performances and findings regarding dark pool performance. Dr. Maclin followed with an introduction to dynamic portfolio analysis and discussion of the dynamic portfolio framework

During the evening session Dr. Kolm presented the stochastic Linear Quadratic Gaussian (LQG) regulator, which is probably the most fundamental optimal control problem, and showed how to do dynamic portfolio analysis with transaction costs. The event ended with Joseph Cerniglia, CFA, senior investment manager at Aberdeen Asset Management, on a discussion on factor-based quantitative trading strategies and back-testing methodologies.

While the event concluded without a major hiccup, some attendants complained about name tags with wrong affiliations, lack of event staff, drinks that quickly ran out, and other inconveniences.

Several Quantnet attendants express disappointment with a lack of hands on, practical experience from the event. While they agree the materials are interesting, many feel it is simply a series of lectures, instead of a “workshop” which would benefit them more if there are citation sessions, coding samples so they can put to use what they just learned.

The cost of attending this two-day workshop runs from $900 for professionals to $550 for academics/students with group and affiliate discounts. The organizers also provided limited $99 registrations for full-time students studying in MFE programs.

The workshop’s website lists IAFE, QWFAFEW, SQA, Axioma, Quant Network as partners of this year’s event.

Billy Diomis contributed reporting and photo from NYC.

NYU-Poly MFE students working in their computer lab

This is series of submissions by students in the NYU-Poly MFE program who share their days. Each day is written by a different student. SHARE YOUR STORY

8:30am – Ring…Snooze…Ring…Snooze…Ring…Alright, alright am up!

8:30-10:00am – Check mail, messages and get ready for the day ahead.

10-10:30am – On the subway ride to school, I read the Wall Street Journal, have to know what’s happening in the world…

10:30-1:30pm – I have my teaching assistant hours to answer questions students might have for the courses in Advanced Financial Econometrics and Risk Management & Asset Pricing.

1:30-1:55pm – Squeeze a quick lunch in between.

1:55-4:00pm – Head to the library to complete assignments and projects. It never ends!!

4:00-4:30pm – Back to the department; am part of the student club (Financial Engineering Association). Meet-up with fellow members to discuss our agenda for the next week, including an upcoming social for the department and a lecture by a trader from the industry.

4:30-5:30pm – Meeting with the members of the trading club. We discuss trading strategies and implications of the present macro-economic environment.

5:30-7:00pm – Back to the library to finish up the assignment for the class later on. Its crunch time… Bring it on!!

7:00-9:00pm – Phew!! Complete just in time. Off to attend the class in Derivative Securities. Come out realizing I still don’t know squat!

9:00-9:45pm – On the way back home, continue reading the Wall Street Journal. Although this time, it’s reading for leisure.

9:45-10:00pm – Quickly review the material covered in class and make a mental note of what I didn’t understand, which ends up being most of it.

10:00-10:45pm – Eat dinner and chill out with flat-mates.

10:45-1:30am – Read up on today’s class and hunt for jobs online. Damn! The economy sucks!

1:30am – Ahhh!! My bed…Zzzzzz

I definitely need to fit in some sort of physical activity to my day. The key to learning from any degree is to be consistent. Have to follow some kind of routine, just keep at it and ask for help if you need it.

ARE YOU A CURRENT MFE STUDENT? Tell us what your day is like.! Email your story to editor@quantnet.com or click HERE

Baruch MFE students in their Quantlab

This is series of submissions by students in the Baruch MFE program who share their days. Each day is written by a different student. SHARE YOUR STORY

2:00 pm – Wake up, bright eyed and busy tailed. Check emails and text messages from my iphone in bed. Check for new course materials/discussions on Quantnet, read Drudge Report and WSJ headlines, check in with teammates, quickly check the market.

2:30 pm – Find out our assignment, due tonight at 6pm, is not yet complete and our assignment, due yesterday, is not yet submitted. Codes required for tonight’s quiz are not yet working.

2:32 pm – Panic.

2:35 pm – Crack open my laptop and frantically start peering over thousands of lines of code.

2:36 pm – Simultaneously open Hulls Futures, Options and Other Derivatives, completing last nights homework via IM

2:36 pm – Curse C++, wish I had written more intelligent code and had commented more.

3:30 pm – Panic.

3:55 pm – Bug is located! Note to self: always put braces around if statements!!

4:00 pm – Hop in the shower and get ready for class.

5:30 pm – Walking to school for 6pm class. Talk to my dad on the way to let him know how yesterdays interview went.

8:30 pm – Meet up with the Baruch trading team to prepare for the RITC competition in February

10:00 pm – Continue working on homework in the Quant Lab.

1:00 am – Cordially escorted outside of the building by Campus Security.

1:15 am – The McDonald’s on 28th and Park has been renovated. Free wifi, tons of outlets, 24 hours and food – score! 4 of us relocate there and continue working

3:30 am – The crowd is thinning out. Start heading home. The city is always peaceful this time of night.

4:00 am – Drop by the bank, post office and the 24 hour Duane Reade. Crack open the laptop and continue coding while waiting for my prescriptions. No wifi here but an outlet works. 

6:00 am – Email team mates my work – they will continue on where I’ve left off. Instant message some friends who have just popped up online.

6:30 am – The sun is starting to peek through the curtains. Take 2000UI Vitamin D. Time to go to sleep.

6:30 am – Alarm goes off. Time to get up for work. I am doing an internship at a private equity/hedge fund in New York. It’s around a 40-50 min train ride from New York City. Hit the snooze button.

6:45 am – Alarm goes off again. Get up and freshen up and get ready for work.

7:40 am – Walking to the subway and realize I don’t have enough time to get to Grand Central to get the train, so I waive over to a cab and have it drop me off at Grand Central.

7:50 am – Get to Grand central and run to Track 18

7:56 am – Off to White plains. Reading the WSJ on the train and listening to Bloomberg podcast.

9:00 am – Walk to the downtown office.

9:05 am – Start my work. Check how the Asian markets did and the issues with Ireland. Check for weekly analyst research from my two subscriptions. Open up Interactive broker’s trader workstation and the Excel DDE interface. Make sure everything is running fine.

9:20 am – Meeting with my boss and go over some of the work I did the night before. We discuss in the conference hall possible ways to interpret the data and charts I developed the day before and how we can set up some effective spread strategies using VIX futures.

10:30am – Get assignments from my boss to work on setting up a calendar spread strategy on VIX futures. Need to pull historical data and set up several graphs for the trader.

12:30 – Go over to Panera for an Italian sandwich while looking over the spread graphs for Cash/Dec, and Dec/Jan.

1:00 pm – Open up the Interactive Broker’s API workbook and start reading, since my next project is to develop an electronic eye on specific spreads that will aide in algorithmic trading models that will be built to go on the IB API.

2:00 pm – Talk to my manager about the next week and what I would be doing since we are moving to Mid-town new York office in a week. I am told I have the week off and I would be working out of the New York office from now on. Only 10 minute walk from my apartment!

4:00 pm – Head over to the train station to go to school. Start working on Numerical Methods C++ homework in the train.

5:00pm – Walk over to Baruch College and go to the QuantLab to hang out with my classmates before class.

6:00pm – Find out today is Object Oriented Programming in Finance instead of Stochastic Processes. SP professor has gone to a conference in San Francisco. Head over to the computer lab instead of the normal class.

8:30pm – After class head back to the lab to play some trading games with my classmates.

9:45 pm – Worked on Numerical Methods in Finance homework. Got assistance on it from one of my teammates. Switched over after a while to catch up on Stochastic Processes course.

12:00 pm – Read about spread trading in the futures market.

12:30 – Log in to my remote access workplace and get onto my work computer from my laptop. Play around with the API tweaking some existing algorithms built by the firm to understand the interface and the code structures so I can start building my project.

1: 30 am – Watch The Office and an old episode of Seinfeld online.

Fell asleep at some point while watching TV shows.

9:30am – I usually got up late.

10:00-11:30am – MFE study involves finance, math and programming. In the morning, I spent time to learn programming if there’s no urgent homework. Recently I was reading two books “Effective C++” by Scoot Meyers and “Inside the C++ Object Model” by Stanley Lippman. The second book is worth reading. Author illustrates C++ features from C++ compiler implementation.

11:30am – Our director helped us set up a Bloomberg onsite training at 12:30pm. Took train to Bloomberg Tower building at 731 Lexington Ave. BTW, other than training, the more often activities are campus presentation. Every week a few financial industry practioners are invited to campus. They introduce their own research topics or quant career. It’s pretty good chance for students to network with senior quants in Wall Street.

12:30pm – Our classmates were gathering at the hallway of Bloomberg building when I arrived there. We’re allowed to enter building after waiting for a while.

1:00- 3:00pm – The training took 2 hours. In the first hour, the instructor made a general introduction, such as how to find economic statistics and price, how to find help document. He demonstrated really fast. I can’t keep with his space. I feel comfortable in the next hour when another instructor focused on details. We learned how to export data from Bloomberg terminal to Excel or directly retrieve data from Bloomberg add-ins in Excel. Bloomberg terminal is definitely powerful. It provides option pricing of different models and hedging strategy.

3:00-4:30pm – Went back to Baruch campus. Some classmates like study at Newman Library where it is more quiet. I prefer to stay in QuantLab, a room exclusively designed for the MFE students. My reason is QuantLab has a mini library containing hundreds of quantitative finance books. It’s convenient to find the book I want to read. In additions, recently three Bloomberg terminals were installed on QuantLab for the MFE students to use. Today’s training will help us take advantage of these expensive terminals.

4:30 – 5:45pm – Discussed Numerical Methods for Finance I (MTH 9821) homework with my teamates. All homework are assigned into groups. Each group consists of 3-4 students with different backgrounds. Homework 11 was to price European/American options using finite different methods to solve PDE. We requires to implement different solvers including forward Euler, backward Euler with Cholesky, backward Euler with SOR, Crank-Nicolson with SOR and Crank-Nicolson with projected SOR, then apply them to different finite schemes to compute option values, greeks, approximate errors. We not only considered algorithm implementation itself but also considered how to efficiently output data to .csv file and into specified spreadsheet. Programming homework is mainly written by C++ and part of homework employs VBA. Homework always killed a lot of my time.

5:45pm – Dinner.

6 – 9:30pm – Classes are offered in the evening beginning at 6pm. Tonight class was Numerical Methods for Finance I. Firstly we took one hour to do a quiz to verify our homework program. Then, professor taught us how to price Barrier, double Barrier, Bermuda options using finite difference methods. Professor always teaches very quickly. I need to keep focused. Usually professor gives weekly assignment in class or post on website after class. Fortunately, our life will be easy this week. We don’t have assignment because of Thanksgiving.

10:30pm – Got home from campus. I read notes and reference books to figure out stuff that I couldn’t understand in class, then started homework. Since there’s no homework, I decided to focus on Capstone project tutored by the C++ class’ professor who is an experienced practitioner. It’s a good chance to learn more practical stuff. Our project just started and will analyze CDS market using PCA. Yesterday I took the whole night to setup development environment, Linux, Eclipse, Python, R, MySQL. Tonight I read carefully two 30-pages papers from the professor. CDS market looks complicated.

2:00am – Went to bed.

10.30 -11.30 AM – Review Notes from yesterday’s lecture

11.30 AM -12.30 PM – Work on homework

12.30 -1pm – Lunch + WSJ

1-4pm – Start working on Homework again

4-4.30pm Work on interview question

4.30-5.15pm Home to College

5.15-6pm Hang out in Quant Lab (study room for MFE students) discussing homework, playing with Bloomberg terminals.

6-8.30pm – Class

8.30-9.20pm – College to Home

10 -12am – By 10, I am usually done with my dinner. Then if homework due tomorrow is pending then I work on it otherwise I read one of the books listed on Master Reading List for MFE students on Quantnet.

12-12.30am – Talk to family and friends in home country

12.30-1.30am Watch markets in home country. I usually have long term positions there.

1.30-3am – I am currently working on a project for a trading company. I usually spend 1.30 hrs on it.

3-10.30am Sleeping. Do not disturb

ARE YOU A CURRENT MFE STUDENT? Tell us what your day is like.! Email your story to editor@quantnet.com or click HERE

Profile of Tim Weithers on the Chicago MSFM’s faculty page as of Nov 21, 2010 (Source: Googlecache)

We wonder whether this departure is symptomatic of some of the turbulence that has affected the Chicago program in recent months. We published a piece on our site a few months back informing readers that the founding director – Niels Nygaard – was stepping aside. Not only was a new director appointed but the new position of executive director created and filled. Like other MFE programs in the current downturn, Chicago’s MSFM has had to contend with sagging placement rates; as a concomitant there has been increasing dissatisfaction from the students who, wanting value for their dollar in these difficult times, have been demanding both more strenuous placement efforts and better taught courses. The departure of Tim Weithers may, however, have nothing to do with the makeover of the Chicago program and any possible connection is strictly conjecture on our part.

Tim Weithers was co-teaching the Foreign Exchange course (Finmath 37300) with Al Kanzler. At the start of the third lecture, students were informed that Weithers had left and the remainder of the course would be taught by Al Kanzler alone. Seemingly no explanation was given. At least some of the students were pleased with Weithers’ departure as it would appear there had been numerous complaints about him in the past.

This year Gregory Lawler will be teaching stochastic calculus, one of the courses most complained about in years past.

CMU MSCF students at the 55 Broad Street campus

This is series of submissions by students in the CMU MSCF program who share their days. Each day is written by a different student. SHARE YOUR STORY

7:30am – I am NOT a morning person but I am trying to be one. My schedule would be so much different if I wrote this few weeks ago.

8:00am – Started working on my Fixed Income homework. It will be due in two days and I heard that it is super very long, so I’d better not wait until the last day. I like to wait until the very last day to do my homework because it’s when I work most productively. Also, that’s how I get time to do other stuff that I want to. I used to do my homework as soon as I can and check/discuss with classmates over the week about it. Eventually, I found that homework was controlling my time.

11:00am – I got bored, so I went to WSJ, Bloomberg & SeekingAlpha to check the news. I normally do not have time to read all the articles that I want to. Also, I know that I would not remember the details anyway, so I just skimmed over the titles and read those articles that sounds very interesting.

12:00pm – Lunch & some music

12:30pm – Back to my homework. It normally takes me 7-8 hours to finish the homework of each class. I like this mini-semester better than the previous one since classes covers totally new materials, making me feel that I am learning a lot more. 

2-3:30pm – Read an Options book that was recommended by a second year in one of our interview preparation sessions. One thing I like about CMU is the coopertation between first year and second years students. Ever since I got into the program, the second year students have been trying to be helpful. I know that they are super busy, but we (the newbies) know that whenever we need help with anything, we can go ask them easily. However, they are graduating soon, so we will be on our own from next semester. 

3:30-4:15pm – Going to school

4:15-5:30pm – Discussing homework and/or hanging out with classmates. The small campus makes us so close to each other. We have three classrooms, 3 study-rooms, 1 leisure room & dining area. A lot of people just leave their computers, books & notes on campus. It does not feel like a normal campus, but it is kind of cool that the whole area is dedicated to MSCF students, so we can do whatever we want with it.

5:30-8:30pm – class time. It is hard to be focused in 3 hours, but I attempt to because it’s most convenient to ask questions in class and it would take a lot more time to re-watch the video online (all lectures in the program are recorded and available students any time)

8:30-9:30pm – going home

10:00-11:30pm – job-related things. Read the news, do some brainteasers & read finance books.

11:30pm – go to bed.

Since August when I started my Computational Finance program at Carnegie Mellon University’s New York campus, life has not been the same. 24 hours a day just does not seem to be enough.

2AM – Went to bed.

10AM – Woke up. I got to get my 8-hour sleep a day to push my body through the long day ahead. Otherwise, I would feel tired during those long evening classes.

My breakfast consists entirely of fresh fruits I usually cook some simple lunch at home because there aren’t any decent eating places near my place in Jersey City, but there are many nice eating and affordable places near CMU’s downtown campus.

11AM-Noon – Spent the first hour checking emails and reading Wall Street Journal. CMU subscribed hard-copy of WSJ to us, but I prefer reading the online version. As part of the subscription, we have username/password to access WSJ online.

Noon-12:30PM – Spend the next half hour making some trades on CMU’s simulated trading competition (lasting about 20 days during the month of  Nov). This is an internal competition sponsored by Deutsche Bank. There will be a dinner hosted by DB in Jan to conferring the prize to winners. The prize is not much in monetary terms, but lots of students find it a compelling way to prove to future employers that they are interested in trading.

You can see the pictures of the last competition’s reception held in Jan 6th, 2010

12:30-2PM – Watched the remaining recorded lecture on Multi-Period Asset Pricing. Since CMU uses a tele-conference system to broadcast the lectures to the other campus, the recordings will be available online immediately. You can watch the video feed in real time as you sit in either campus. Many students watch the recorded lecture more than one time to understand the material well. It should be noted that previous year’s lectures are not accessible.

So far, I love every class that is challenging because I thought the whole purpose I came to CMU to study is to learn the financial models and quantitative methods, such as those found in stochastic calculus or multi-period asset pricing.

2-4PM – Reviewed and practiced the sample questions for the Fixed-Income mid-term later in the day. For most courses, the professor will post some past-year questions for review. Of course, there will be “new” questions in actual exam.

4-5PM – Get ready to travel to CMU’s classroom at 55 Broad Street in downtown NYC.

Door to door, it takes around 40 minutes to reach the class from Jersey City via PATH. I usually read some e-books on my iPod Touch on the train or just listen to music to relax.

5:30-7PM – Took mid-term test

7:30-8:30PM – Continued the lecture after the test. The classes in NYC campus are all from 5:30-8:30PM. Only a few classes have mid-term tests. All classes have weekly homework.

9:00PM – Got home.

10PM – Started working on homework and assignments that are due in a day or two. All HWs are posted by professors. Some HWs allow work in group, but most of the HWs are individual-based though the professors encourage collaborations among students.

1AM – Checked financial news update from Asia on my iPod Touch on my bed before I go to sleep, and get my 8-hour sleep, for another day.

The above is just a typical day on my life as a CMU MSCF student. While not busy studying, I attend MSCF Speaker Series and company presentation. Networking events as well as the Trek to companies in NYC have been very useful in preparing me for my internship search.

ARE YOU A CURRENT MFE STUDENT? Tell us what your day is like.! Email your story to editor@quantnet.com or click HERE