Hello everyone. This topic has heated up since I left for vacation. (I've just come back from 4 days at a health spa. It was great fun in Arizona.) Regarding the quote about extremism in the defense of liberty, it was Barry Goldwater who said that.Re: Hamilton's questions re: flexibility within a big organization...here goes. I've been thinking about this a lot so it's a timely question for me.I agree that many companies are "results oriented". This makes it hard to justify research that will take a long time until it bears fruit (i.e. for the company -- making money is "bearing fruit"). I also think that being especially insightful in a research sense (always asking how things work and why) may be a detriment in a big organization. Why? Because they are results-oriented. That's not a bad thing unless you want more time to ponder than the company can "afford".I'm trying to develop my career around things that are intellectually challenging to research and develop. For me, that means combining research with neat consulting work, speaking opportunities, offering executive seminars about what I have researched, etc. That assumes, however, that the research is "added value" to someone. My personal preference is to combine theory and practice in solving problems. I acknowledge that this is different from pure theory and that's fine with me. Networking with a mixed group also makes a big difference. I'm a member of a quant. finance group called QWAFAFEW. See http://www.qwafafew.org. Their membership is made up of academics and practitioners. Tonight, I went to a talk about backtesting and data problems. The speaker was the head of quant. technology for CitiAsset Management.

Barry Goldwater indeed spoke those words at the 1964 Republican National Convention, but Karl Hess wrote them. The actual line begins with "Let me remind you that. . ." which suggests Hess may have been quoting a familiar saying (he died in 1994 without settling that issue). Personally, I prefer Lyndon Johnson's reply, "Yes, Barry, but extremism in pursuit of the presidency is an unpardonable vice." Everyone remembers the ringing words, but everyone voted for the plain common sense.Both Goldwater and Johnson were bandying words, of course. Someone who really wrestled with the question of extremism was Malcom X, who debated the affirmative (i.e. pro-Goldwater) position in the Oxford Union debate later in 1964. He traced the sentiment in American politics back to Patrick Henry, but ultimately fudged by defending a qualified extremism (it had to be "justified" and "in defense of HUMAN justice," emphasis his) which, of course, means it is no longer extremism.

Susan wrote:"I agree that many companies are "results oriented". This makes it hard to justify research that will take a long time until it bears fruit (i.e. for the company -- making money is "bearing fruit")"Let’s face it, most people in this business are Greedy. But there is short-term-greedy and long-term-greedy. To many in this business are short-term-greedy. The few Wall Street firms/people that are known to be long-term-greedy are typically putting more money into research. High turnover of people and bonuses typically based on how much you made this year are naturally one of the reasons.

Do the 5% have an incentive to minimize the qualifying faculty so as to increase their relative standing? >>It is not so much about relative standing with respect to colleagues as about research time. The more colleagues you have that are being 'punished' with teaching obligations,the less teaching you need to do yourself. That means research time, more publications, less teaching etc etc. I am not against research output measurement, but transparency as far as the criteria are concerned should be rule n. 1.PS The 5% do have their own Journals, Aaron!

The more colleagues you have that are being 'punished' with teaching obligations,the less teaching you need to do yourself. >>How many contact hours per academic year would you call punishment?

How many contact hours per academic year would you call punishment? >>Those who do not meet the standards need to do twice as much as those who do. About 350 hours, I believe.

Dear all,I'm a member of this forum since less then one week and I'm suprised from the quality of the discussion. I'm working since one year for RiskLab, an Institute of the Swiss Federal Institute of Technology in Zurich, which is sponsored by two of the most important banks in the world (Credit Suisse Group and UBS) and one the most important reinsurance (Swiss Re). Our sponsors as well as academic partners propose to you projects, which are of relevance for the academic world but also for the application in Risk Management. I'm currently working on Credit Risk, dealing with retail portfolios. The approach I proposed and analyzed for solving the problem of modeling the default probability is quite attractive from the point of view of the mathematical properties, as well as for the intuition. Nevertheless the lack of information, the bad quality of the data set make it very difficult to test the model in a satisfactory way. Academic research cannot naturally be implemented 1:1 in the practice, but the distance between Universities and the "real world" would be less if people who are making research have more experience about the real world. The same is true, if portfolio managers or risk managers try to follows the last findings in the academic research. I think that one solution would be that a research programme is divided in two big parts, the first one where researchers deal for some periods (one-two years) with financial instrument and the second one where they try to learn about the technique for modeling the risk. This would increment the ability of researchers to find solution which are of relevance for the practice, and thus accelerate the decision process within the financial institution.Best reagrds,Enrico

Enrico, I'm very interested in doing research in the credit risk area as well. I'd like to learn more about what you are doing and Risk Lab in general. I agree with you about doing something on a "collaborative", joint academic-industry basis.Susan

Hi Susan.About my project you can find on my personal homepage http://www.math.ethz.ch/~degiorgi the slides I've used for the RiskLab Workshopo of June 7. If you are interested in something more recent, I can send you the slides I've used for a presentation in Zuirch last August 21. I hope that the preprint version of the paper will be available as soon as possible. My project was proposed by the Credit Suisse Group. Currently the bank uses a modified version of CreditRisk+, which is not very satisfactory for retail portfolios, since the different factor, which affect the ability to pax the interest rate on an oustanding mortgage, are not really captured. I proposed an intensity based approach, and I've model the conditional intensity process by using a multiplicative model, analog to the Cox model, but the reletionship is not assumed to be linear. This implies that the estimation requires some smoothing technique, to find the relationship.RiskLab have currently other project on Credit Risk. One project try to model the Credit Risk together with the Market Risk using a firm's value approach. Prof.McNeil (the same who give the congress on application of SPlus next week in New York) and Prof. Frei collabotare with RiskLab, and have recently written a paper on Modeling Credit Risk (see Alex McNeil Homepage). They show that the methodology proposed by JPMorgan (CreditMetrics) and the one proposed by Credit Suiss First Boston (CreditRisk+) are essentially the same (a paper on the subject also appeared on the Journal of Banking and Finance, by M. Crouhy, D. Galai, R. Mark (2000, Journal of Banking and Finance 24, pp.57-117)). You find the link to all these papers on my Credit Risk Home Page (see my Personal Home Page).Other information are available on the RiskLab homepage RiskLab. If you special questions about my project, please contact me.Best Regards,Enrico

This is great. Thank you. Unfortunately, the S-Plus seminar on risk management was canceled. Re: credit risk data, are you getting this from Credit Suisse or a general database (recovery rates, default rates, etc.)?

Hi Susan. The data set is unfortunately confidential. I get it from Credit Suisse Group.Best Regards,Enrico

Thanks. Getting good data is a big deal. In fact, re: publications, my understanding is that some papers are published because the data set is unique. Of course, that makes it hard to verify results, i.e. peer review.

You are right about receiving data set. But even if you receive a data set, specially about retail loans, there is a lot of difficulties that has to be solved, before you can test one model. Without a good collaboration with the industry it is quite impossible to find out, wheter the data set can be used or not. In fact my main problem is that my data set contains some stange behaviour, which is due to the way, data has been collected!! I thing that the goal of an academic research is to give new approaches, for example for modelling the market risk. My model for example is nothing absolutely new from the mathematical point of view, but the application on retail loan portfolios and the estimation methodology that has been applied in my work, maybe are new. My goal is to explain how you can model default risk and how you can estimate the model to get good results. It is clear that the data are not available to other researchers and therefore my results cannot be verified. But nevertheless, one methodology, one solution has been proposed and people are interested in, can use the same tool on the own data set: the methodology will be the same, maybe with little modification! Naturally Credit Suisse Group will be moreover interested on the specific results, not only on the methodology.Best regards.Enrico

You bring up an interesting issue. If something is of help to a limited audience, will academic researchers pursue it? Anyone have thoughts?

Susan, EnricoSeems like you're discussing something even bigger than research academia vis a vis financial markets per se... more importantly, its the age-old question of how much of what you learn in the academe do you actually apply in practice?Having experience in both the academe and now just starting in applied financial markets, it seems really tough for quants to apply what they learn in the academe and put them to practice. (Hmm, well, maybe I'm just speaking for myself...) Most of the quant methods end up in backoffice activities: risk management, research, consultancy... but when it comes to front office activities it seems like very little quant is applied...(after all, a firm's economic goal is to mazimize profit...)I am not sure how it is like in developed markets, but for developing markets where I am working with, I see VERY little use of quant methods in front office activities. Everything just seems to boil down to local supply and demand conditions...of course it does get a bit disappointing for me...Enrico, interesting website you've got. Susan, any quant orgs you're familiar with in the Asian region?