SERVING THE QUANTITATIVE FINANCE COMMUNITY

 
User avatar
lehalle
Posts: 50
Joined: April 19th, 2006, 7:13 am

1/3 quants are french ?

May 24th, 2007, 7:43 am

QuoteOriginally posted by: quantwanabeIf you have a nobel prize winner as a teacher, will you win the nobel prize as well???If your teacher is a Nobel price winner in physics, you will have a better understanding of physics that one with a monkey as teacher...Moreover, teachers always selecte their students. You cannot send your mail to Nicole Elkaroui (or Gilles Pages, or Marc Yor ; they are actually 3 to run this DEA) and come to the course the week after, you have to send your CV, some sample of your work, and have an interview.Of course you can learn alone, but it's much more harder.Just put the name of academics that you respect in the math genealogy database and check their teachers, you will not be surprise. By the way, it seems that Nicole's PhD director was Jacques Neveu, but I do not know the name of the PhD director of Neveu, any guesses?
 
User avatar
miquant
Posts: 36
Joined: March 16th, 2007, 8:07 pm

1/3 quants are french ?

May 25th, 2007, 3:17 am

I guess i am in the right place to give my opinion about french quants, the real value of masters degree in France because i am french, about to graduate in one of those masters (actually the ¨opponent¨ of El Karoui´s one) and i have been spending time with students from all kind of french schools and studies this year.First, i have to admit (and i swear it is not chauvinism like most of you guys can think) that french mathematics school is certainly one of the best in the worl, with the Eastern'Europe ones, and it has a lot to do with the massive number of french guys in the industry right now.I have been living in NY and was about to enter in Colombia or NYU finacial engineering masters last year (i am phD in applied maths, but not finance from 2003), and i have been doing my research about teachers and courses given in NY and in Paris. It is evident that those mastes sounds alike on the paper but if you ever have the experience of assisting in a french and a US course you will undoubdly see a difference in the way mathematics are taught.I am convinced that the french masters have more difficult mathematics upgoing when studying Fin. Eng. than US or British students, i guess we are a little weak at computer programming because lots of them don´t learn how to code before very late (like last year), but the most important part is that we have some of the best teachers and are able to get to choose in a large range of classes very market'oriented.Well ... i may be wrong on some points because i don´t know everything about what´s taught around the world, but i guess french quants like to work with people from the same school because they know there will be NO surprise. El Karoui´s master has become a gold line on a resume and i have to admit that best quant probably come from this master because the majority come from Polytechnique which represent a large part of the most talented french in mathematics (except from students from the Ecole Normale Superieur de Mathematiques). Also it must be said that lots of others good schools teach finance in the last year and permit to get a master as polytechnic do. And i repeat i am not from these schools, i am from University ...
 
User avatar
Tadragh1
Posts: 115
Joined: July 3rd, 2006, 10:06 am

1/3 quants are french ?

May 25th, 2007, 6:45 am

Miquant, I have heard many opinions of French quants (on general, as you say they are in general very well mathematically educated) being very theoretical, sometimes to the point of being too theoretical. It seems you actually confirm it, when you say about programming skills. What do you think?
 
User avatar
lehalle
Posts: 50
Joined: April 19th, 2006, 7:13 am

1/3 quants are french ?

May 25th, 2007, 7:18 am

QuoteOriginally posted by: miquantI guess i am in the right place to give my opinion about french quants, the real value of masters degree in France because i am french, about to graduate in one of those masters (actually the ¨opponent¨ of El Karoui´s one) and i have been spending time with students from all kind of french schools and studies this year.You are wrong seeing those two DEA as "opponents"... Explaining the success of french people in mathematical finance (I do not say that french are "the best" in this field, but they are relatively good in it) is difficult.My opinion is that some of academics who founded and develop martingale theory and stochastic processes are french (as Paul-André Meyer, Jacques Neveu, Marc Yor, Jean Jacod, etc) and their students which now work in mathematical finance stay close to their laboratories. The LPMA is a very good example of such a mixture : probability theorists and finance academics working at 10 meters of distance. A lot of academic advances in PDE come from french people too.I think this is the cause of the success of french mathematical finance.I do not say that french are "better" that others, but that the mixing of theoretical and applied academics in the same labs could be a good explaination of french efficiency in the field of math finance.Another more generic explaination can be found : french people really like abstraction. All french people studied philosophy at school (it is mandatory for all students), latin language is widely teached, set theory has long been teached to all 10 years old french children.This is good to produce mathematicians (see the # of field medals relatively to population size), and somehow good to produce quant, because math finance is really an "abstract application" of mathematics : who have ever seen or touched the N-dimensional stochastic diffusion that support assets moves? this is not like for instance PDE governing the molecular behaviour of steel during the galvanization process...The counterpart of the is that it's very difficult to find a math finance book written by french which does not begin with "Let have a Polish space"....
 
User avatar
miquant
Posts: 36
Joined: March 16th, 2007, 8:07 pm

1/3 quants are french ?

May 25th, 2007, 8:41 pm

QuoteOriginally posted by: Tadragh1Miquant, I have heard many opinions of French quants (on general, as you say they are in general very well mathematically educated) being very theoretical, sometimes to the point of being too theoretical. It seems you actually confirm it, when you say about programming skills. What do you think? I guess this is true ... from my own opinion, frnehc university and Schools should have more intensive programming classes, especially about C++ at an early stage of studies because it is hard to have to tackle with programming from nothing when you find yourself in a Master like El Karoui or others.About the ¨opponents¨ DEA, i was trying to find a bette word but couldn't with my lack of english vocabulary, i wanted to say that the other DEA is also devoted to quant jobs, with teachers in commun, ... different timing because classes end 2 months earlier in El Karoui master. But anyway ... i did not intend to say that students or teachers are opponents.Your analysis is right about laboratories and theoretical approach of the french shool.About french books not begging with polish space, there are more and more now, like Lamberton one, or Meyre newly published book about stochastic calculus applied to finance.
 
User avatar
CrazyAsian
Posts: 23
Joined: September 29th, 2002, 4:56 pm

1/3 quants are french ?

May 25th, 2007, 9:00 pm

QuoteSeveral highly placed Silicon Valley first generation executives lament that their children want to become doctors, lawyers, investment bankers rather than go into engineering or other sciences.I wouldn't doubt that but realize that most Silicon Valley execs have a skewed view of the world. Many of the tech execs I've talked to thinkit is rather easy to find funding, start a tech company, and then sell it for millions. They can't understand why the rest of the world would doanything else.But I doubt that a Silicon Valley exec would want their kids to become civil engineers so the whole engineering bias has limits...
 
User avatar
NorthernJohn
Posts: 632
Joined: June 2nd, 2003, 9:07 am

1/3 quants are french ?

May 27th, 2007, 12:02 pm

QuoteOriginally posted by: pb273QuoteOriginally posted by: NorthernJohnQuoteOriginally posted by: pb273I have been interviewing many candidates for quite a few quant roles (for New York) for the last 6 months, and given what I get, I can tell you that out of every 20 guys, 12 are Chinese, 4 are Indians, 2 are Russians/East Europeans and only 1 is French and the remaining 1 is rest of world incl Brits & Americans.That is very different to my experience. In my last few banks, the teams have been rather French heavy. The curent team is 75% French, actually, and this is not a European bank.They are definitely overrepresened, but this is not a huge surprise. In England, the brightest people tend to do science degrees. In France, they tend to be more likely to do a financial course. As quans, in general, need to be both bright and tutored in the subject, this gives them an edge.So, most likely you have been recruited for just being a french by banks already filled with french - but's thats just a conditional probability.What are you talking about? I am not French, and I am not a quant.
Last edited by NorthernJohn on May 26th, 2007, 10:00 pm, edited 1 time in total.
 
User avatar
crowlogic
Posts: 658
Joined: May 22nd, 2005, 6:47 pm

1/3 quants are french ?

June 18th, 2007, 11:45 pm

QuoteOriginally posted by: miquantQuoteOriginally posted by: Tadragh1Miquant, I have heard many opinions of French quants (on general, as you say they are in general very well mathematically educated) being very theoretical, sometimes to the point of being too theoretical. It seems you actually confirm it, when you say about programming skills. What do you think? I guess this is true ... from my own opinion, frnehc university and Schools should have more intensive programming classes, especially about C++ at an early stage of studies because it is hard to have to tackle with programming from nothing when you find yourself in a Master like El Karoui or others.About the ¨opponents¨ DEA, i was trying to find a bette word but couldn't with my lack of english vocabulary, i wanted to say that the other DEA is also devoted to quant jobs, with teachers in commun, ... different timing because classes end 2 months earlier in El Karoui master. But anyway ... i did not intend to say that students or teachers are opponents.Your analysis is right about laboratories and theoretical approach of the french shool.About french books not begging with polish space, there are more and more now, like Lamberton one, or Meyre newly published book about stochastic calculus applied to finance.Whats wrong with polish space?
 
User avatar
Polysena
Posts: 230
Joined: January 25th, 2007, 10:06 am

1/3 quants are french ?

June 19th, 2007, 9:01 am

Perhaps the right phrasing should be... does France export 1/3 of the quants it produces? Then the answer might be yes... (big issue now discussed hotly in France--- brain drain) There are a few things that may support this:1. academia prospective for whizz are not hyper appealing in France (pay& career prospects)2. those who can convert convert to "quant" and leave or stay in a Business School ( but you have to have a talent for that not all endowed math person will make a good quant).. Some people are better at staying in maths... because of top talent in maths or /& lack of talent for finance 3. those who are at the entry point make new choices ( selecting a Financial Engineering path rather than a pure maths path). And hope that the fame secured by Nicole ElKaroui and the generation of succesful student she produced will help them install themselves ( but as for HF caveats) past success is not full predictor of future successes.4. Given a certain level of quality, the homogenity of cultural background plays a role... if I am Chinese I might prefer to deal with people with whom I can also converse in my native language, idem for all non native speakers- given that one might also prefer to relate to people that have had say Neveu's books as babyfeed than those who were Kolmogorovized..again: just for ease of communication of non-technical stuff.5. Diversity may bring however a plus - because for instance the probabilist school in Russia and French have different flavors (underlying theorems are identical) but then it is the way you do, talk and daily do math business.. having all sorts of cultures and anglosaxons is alos important because the whole business is conducted in English at the end of the day..Finally in terms of exportability:Why are there more Chinese than say others in Us universities- education-abroad policies, money put into sending students abroad, return-home packages and convenants.. plus willingness to delocate from Home vs the expected advantages vary with each country?Just a few thoughts....http://www.maths-fi.com/Gilles-Pages-Ar ... rveaux.pdf
Last edited by Polysena on July 1st, 2007, 10:00 pm, edited 1 time in total.
 
User avatar
Cuchulainn
Posts: 63371
Joined: July 16th, 2004, 7:38 am
Location: Amsterdam
Contact:

1/3 quants are french ?

June 19th, 2007, 10:58 am

QuoteWhats wrong with polish space?No metric space? No Cauchy sequences? Can I program them in C++?
Last edited by Cuchulainn on June 18th, 2007, 10:00 pm, edited 1 time in total.
 
User avatar
crowlogic
Posts: 658
Joined: May 22nd, 2005, 6:47 pm

1/3 quants are french ?

June 19th, 2007, 4:24 pm

.
Last edited by crowlogic on July 22nd, 2008, 10:00 pm, edited 1 time in total.
 
User avatar
Cuchulainn
Posts: 63371
Joined: July 16th, 2004, 7:38 am
Location: Amsterdam
Contact:

1/3 quants are french ?

June 19th, 2007, 5:45 pm

QuoteOriginally posted by: crowlogicQuoteOriginally posted by: CuchulainnQuoteWhats wrong with polish space?No metric space? No Cauchy sequences? Can I program them in C++?Polish spaces are homeomorphic to metric spaces, no?The space induced by the metric for x,y in (0,+inf)d(x,y)=|1/x-1/y|+|x,y|is a polish space.I dont see why you couldn't code them in c++, but a language with more symbolic support would probably be better as an initial step, maybe some code generation.Edit: plus, not minusThe maths must be characterised by some kind of prescription so that we can compute a solution. That's the usefulness of Cauchy sequences. Quoted(x,y)=|1/x-1/y|+|x,y|What is the geometric interpretation of this metric?
Last edited by Cuchulainn on June 18th, 2007, 10:00 pm, edited 1 time in total.
 
User avatar
crowlogic
Posts: 658
Joined: May 22nd, 2005, 6:47 pm

1/3 quants are french ?

June 19th, 2007, 5:58 pm

QuoteOriginally posted by: CuchulainnQuoteOriginally posted by: crowlogicQuoteOriginally posted by: CuchulainnQuoteWhats wrong with polish space?No metric space? No Cauchy sequences? Can I program them in C++?Polish spaces are homeomorphic to metric spaces, no?The space induced by the metric for x,y in (0,+inf)d(x,y)=|1/x-1/y|+|x,y|is a polish space.I dont see why you couldn't code them in c++, but a language with more symbolic support would probably be better as an initial step, maybe some code generation.Edit: plus, not minusThe maths must be characterised by some kind of prescription so that we can compute a solution. That's the usefulness of Cauchy sequences.Right, the solution in this case is more like a recursive likelihood procedure that determines time-varying exponential decay parameters, it's an "inferential process".QuoteQuoted(x,y)=|1/x-1/y|+|x,y|What is the geometric interpretation of this metric?x and y are inter-arrival times between some independent events from a Poisson process which must be >0 because no more than 1 event can occur in an instant of time. You can drop the requirement that x and y be less than infinity since infinity is the event that never occurs in which case the distance between some never occuring event x and some other event (occuring or not) y is infinity. It occurs in the study of finite point processes and compensators.We also havewhich is 0 when epsilon=0
Last edited by crowlogic on June 18th, 2007, 10:00 pm, edited 1 time in total.
 
User avatar
jarael
Posts: 2
Joined: February 22nd, 2007, 9:30 am

1/3 quants are french ?

June 21st, 2007, 8:04 am

I think it's more a case of "where you have one, you have many". The French seem to prefer hiring solely other French speakers, no matter which "nationality" the IB. I actually know of one person for whom this is a recruitment policy.
 
User avatar
jfuqua
Posts: 1255
Joined: July 26th, 2002, 11:41 am

1/3 quants are french ?

June 21st, 2007, 2:03 pm

A probability/finance prof. told me he was amazed at the number of dissertations in finance [and other fields] from non-English countries that were written or at least translated into English, whether they were publically available [e.g. Web] or not. One possible implication is that they feel to get known, they better publish in English. I use to see a few working papers in German and a few in French but I don' think I've seen one in several years. Recently a math finance PhD from Estonia was done in English.
ABOUT WILMOTT

PW by JB

Wilmott.com has been "Serving the Quantitative Finance Community" since 2001. Continued...


Twitter LinkedIn Instagram

JOBS BOARD

JOBS BOARD

Looking for a quant job, risk, algo trading,...? Browse jobs here...


GZIP: On