UCL Financial Computing

[brucebanner]

Hello Folks,

I did the Financial Systems Engineering (UCL-CS SSE: MSc Programmes) at UCL (University College London), which has a suprisingly good financial computing department. See

UCL Partners with SunGard

UCL computer scientist honoured by Cambridge
Financial Services Knowledge Transfer Network
UCL-CS Information
Financial Computing at UCL

for more info. BTW, do you guys have any opinions on UCL's finance computing courses? Many of my friends - who also did the course - now work in investment banks (IBs), but not as quants - grad programme.

I did a Computer Science degree as an undergrad and I'm currently writing financial monitoring software in C++ and I'd like to get into quants. Is there any advice you can offer.

Will a Ph.D. in Financial Computing (http://www.findaphd.com/custadverts/4year/ucl_engd/fc.asp) offer me a route as a quant into a bank?


PS. the forum is very clunky, especially the HUGEEEEE buttons.

 

[DominiConnor]

This course belongs to the very small set that I recommend to no one.
UCL is a top tier place, this is not a top tier course.

It's not "financial computing", it's "lectures we can give that we can sell for 3 times the price than if we called them 'introduction to IT"

Of course none of brucebanner's friends work as quants, it's not a quant course, and teaches little of value to anyone considering this line of work.

If you want an MSc then do an actual maths course, or something involving computers, preferably programming.

For the avoidance of doubt my evaluation of your academic background exponentially declines with the number of "people issues" modules you've done.

...except if you're a security specialist. But not when it's something poked in to allow the slower witted to get an MSc like it is here.
If you want to do security go to Royal Holloway, not UCL.

Any masters course that says that it's aimed at people who don't understand the subject fails to get past my filter.

 

[brucebanner]

Firstly, your message was incredibly rude and harsh. I won't even comment on it.

And you didn't even answer my question - "Will a Ph.D. in Financial Computing (Home) offer me a route as a quant into a bank?"

 

[DominiConnor]

I'm sorry if I offended you, not very sorry of course since my role is to warn people where I can of poor decisions.

Just to reassure you, I'm normally ruder about this course, I wrote the short version.

I don't see any PhD with that sort of title being the ideal route into being a quant. It sets an expectation that your grasp of the relevant maths is at best 2nd tier, also there exists a prejudice that people who do careerist PhD are less smart.

Also if you google my name you will find me repeatedly telling people not to choose "career" PhDs because they do not give the expectation and variance of career outcome that you might think.


If I had to choose such a course for some reason, then UCL is not a bad place to do it. Piotr Karasinksi is a top class guy with both academic and real business experience. Having him as your supervisor would be good.
But I don't think you will get that, for various reasons.

I don't know Nuti's work in any detail, but he has a cool job title and can no doubt tell you seriously valuable things. How much spare time he has to actually share it with you is a big variable. Algotraders are not known for their lazy lifestyle.

We have a problem with "executive PhDs", and we've said unkind things about these at Imperial.
A PhD is a research degree, not a long MSc, and a big reason banks like them is the demonstration of you being able to do hard things without having your hand held.
Thus I am concerned that research is the second bullet point and barely mentioned elsewhere.

They key term is not that it is hard but that you can be independent, but of course both are good.

I can't yet say whether UCL will fall into this hole, the fact that it is based upon their MSc does not help me like it.

I like the idea of apprenticeships, which is what this may actually be, but they are not PhDs.


It could be excellent, they have the components to make the first decent program of this kind. The fact that there isn't one yet should alert you to how hard it is.

 

[Joy Pathak]

I have never really seen a PhD program promoting itself as an industry degree. Well untill now...

As leading banks and funds become more scientific, the demand for excellent PhD students in the above disciplines is soaring.

Almost always top school PhD adcomm will reject your application if they see any sign that you want to pursue anything but academia upon graduation. It is a different thing that many choose to go the industry route upon graduation, but that is usually because they cannot get academia jobs-no publications-etc.

I am very surprised a program is promoting itself as a target doctoral degree to enter industry.

 

[bigbadwolf]

Almost always top school PhD adcomm will reject your application if they see any sign that you want to pursue anything but academia upon graduation. It is a different thing that many choose to go the industry route upon graduation, but that is usually because they cannot get academia jobs-no publications-etc.

If the schools are rejecting applications on this basis, they are being incredibly disingenuous. Everyone knows there aren't enough academic jobs around for the torrent of PhDs being released into the job market. This is a different matter from choosing doctoral students who can hack their way to a decent doctoral thesis.

 

[Joy Pathak]

If the schools are rejecting applications on this basis, they are being incredibly disingenuous. Everyone knows there aren't enough academic jobs around for the torrent of PhDs being released into the job market. This is a different matter from choosing doctoral students who can hack their way to a decent doctoral thesis.

Well I guess it depends on the PhD field. In PhD Finance this is almost a given that the ad-coms dislike applicants who show preference to work in industry after.
I faced similar views by ad-coms during my engineering PhD applications, but I guess it might be unique for the schools I applied to.

PhD Finance from decent schools have 2-3 students who graduate every year if at all. Other scientific fields with more graduates might be more lenient in applications with industry prospectives.

 

[DominiConnor]

At various times and laces it has been common for PhD to be part of a career structure.

This has been true in Chemistry and Engineering. Often it is the case that one needs access to resources that could not possibly be held by a university. A decent % of applicable science PhDs are sponsored.

So why am I so cynical about 'executive PhDs' ?

Partly it is the path of the money.

Executive PhDs seem to have as the primary design goal, making money for the business school. This is different from industrial PhDs in other subjects where the university makes very little out of it.

I'm all for having practitioners come in to help explain the 'real world', but I question whether a senior manager at a bank has the time to be a PhD supervisor. If he has that much time, I question whether he will remain a senior manager....

Banks are really shit at career development, they hire what they want at the price they need to pay, when they need to pay it.
This means their commitment to a process where no result of any kind will be seen for >4 years is going to be unreliable.

You might say "but manufacturing firms are like that now", and I will say "yes".

That's why we see much weaker links between engineering departments and manufacturing firms.


The UCL site has various bank names plastered over it, I will guarantee that out of any 5 banks you pick, two won't be in the same form when you finish this course. That's not because of the recent clusterfucks, it has always been like that, JP Morgan's 'full' name is somethning like JP Morgan Casenove. Chemical Chase Manhattan Bear Strearns...
several of those names are themselves composites.

I'd give it 50/50 that in 5 years 2 out of 5 have merged into a definite entity, or split up, or some combination of the two. P&D has been a client of 4 different banks (including JP M) without changing it's account, I think we're with Santander this week, but that's a variable.


Don't get me wrong, I would love this program to succeed. We need people with deeper, more applicable training. I find it ironic that the problem is money itself being a bad influence.

 

[bigbadwolf]

Banks are really shit at career development, they hire what they want at the price they need to pay, when they need to pay it.
This means their commitment to a process where no result of any kind will be seen for >4 years is going to be unreliable.

You might say "but manufacturing firms are like that now", and I will say "yes".

That's why we see much weaker links between engineering departments and manufacturing firms.

Agreed. The weaker links between engg depts and manufacturing firms has been discussed in some of the British broadsheets in recent weeks.

 

[zeeshan]

This might be a option at UCL http://www.ucl.ac.uk/prospective-students/graduate-study/degrees/pgt/TMSMATSMDL01

 

[Barny]

Dominic - I think the aim for this programme is to get people into the tech divisions at IB's, rather than work as quants. If you look at the syllabus, it includes lots of project management style things and Java programming. As for the PhD, my impression is that you get to potentially work on a decent problem - but actually many of the PhD's are done via another department. For instance, I spoke to the director of the programme, and he mentioned that a guy in engineering working on monte-carlo simulations to some engineering problem adapted them to a finance problem and worked at BNP during his summers. Sounds like a pretty decent deal to me..

 

[Daniel Duffy]

Some thoughts on these courses, seeing my involvement in similar endeavours since 1990..

I had a look at this course. In my opinion it is what you should learn 1) after 10 years of hard-core programming experience or 2) on the job (learning Tools for later).

You learn all this stuff doing, on the job.

I would learn hard maths, data structures, C++, parallel processing, etc.

Another example is the set of courses
http://www-typo3.cs.ucl.ac.uk/students/syllabus/mscsse/gs02_advanced_analysis_and_design/

It is good to do these; I have intimate knowledge of all these IT (i.e.they are not QF-specific) topics (OOA/OOD since 1991). Software has moved on a bit since the heyday of UML and GOF, the world is not OOP alone but is multi-paradigm.. And quant s/w has a different focus; it is more focused on algos rather than on software design. UML is no longer a hot topic. You cannot become a good designer/architect until you have become a good developer.

Actually, OOP is not the most useful model as such for quant software...there is a mismatch, but OOP does play a role.

Last, how to link up OOP and QF???

I tend to agree with Dominic's viewpoint. He has real experience and knows what skills are needed.

 

[jpglotzer]

Does anyone have an opinion about the MSc program posted by zeeshan?

http://www.ucl.ac.uk/prospective-students/graduate-study/degrees/pgt/TMSMATSMDL01

It seems to be pretty thorough on the maths end, and 3 optional modules could be:

Financial Mathematics
Mathematics for Economics
Applied Mathematical Finance

I suppose if I'm aiming to be a quant, I should just do an MFE, but would this course maybe be more flexible with respect to career options? Or would I mostly just be compromising my chances of becoming a quant?

 

[Hua_GeGe]

Hi guys,

I'm also interested in the phd Financial Computing. It seems that the program is not that popular as i can only find few info online. Any comments about the program?

 

[zeeshan]

The course seems interesting but obvious if you search online for information, it is not that recommended.
DominiConnor Which institution would you recommend for a phd in financial computing ( more algorithmic in nature) ?

 

[mat_seymour]

DominiConnor, not wanting to stray too far from the subject of the thread, but I was wondering if I may ask your advice. I have recently graduated with a master of physics degree, and would be looking to become a quant later down the line. In my final year I took a module in "complexity science" and would be interested in further studying this field e.g. Warwick's complexity science doctoral training center. Is this a sought after phd for a quant career path? Or should I stick to research in a more "pure" physics area? If others are interested in discussing this it may be worth starting a new discussion rather than hijacking this one.

 

[TehRaio]

would be interesting to know if the opinion on the financial computing program of ucl/lse has evolved? it's been five years since that thread was started, I understand this forum focuses mostly on american programs but still, anybody?

 

[Invid]

Does anybody have any news about the program? How do you think it will evolve, now that Brexit is in place?

 

[Oll]

So I'm not sure how much help I can be regarding the course, but unfortunately UCL has turned into a profit hungry business. I've been here for the last 7 years, doing undergrad to MSci, and now MSc and it has certainly changed a lot in that time. I helped in enrollment at the start of this year and they've upped the amount of students by around 4000 and now UCL is really struggling for space for its students (true stories of lectures being held in hotel lobbies for Physics).

Anyway, the course seems pretty good and most people tend to get a summer placement I think. It seems like the CS department really looks after its students.

 

[F.Fontana]

I've just graduated from the MSc Computational Finance in UCL. Feel free to ask me specific questions.

Quick review - Since half of the program is composed by elective courses, you can really tailor the structure of the program according to your preferences. This is wonderful if you know what you want. The department offers many courses in machine learning, thus if you want to study both machine learning and quant finance, this place may be very suitable to you. The career service placed me in one of the top investment banks doing applied machine learning and I'm still there.

P.S.: I've had colleagues coming from the aforementioned Financial System Engineering and some are still there as well.