Search results

First one’s free...

Thank you. Dan Stefanica and I have been planning this for more than 5 years. Time to execute.

No charge.

I will soon post some governance-related files.

1. Culture varies widely from area to area. If you're a quant developing a new approach to AML in compliance, you're on virgin snow and likely working alone. If you're an interest rate modeler or PDE person working on option pricing, you're essentially a commodity. Smart is necessary but not...

I concur.

Here is the MS Stuff spreadsheet many have asked about. I usually go through this in my classes. Please do not ask me to annotate it or provide additional details.

I work with these people daily. It’s very technical, but requires a variety of additional skill sets. Banks rely on audit to ensure that technical functions are doing a) what is required by internal policy and procedure and b) what is required by regulators. Thus, technical auditors need to...

@ApolloChariot, under normal circumstances, I might agree with you, but knowing the details and the people, I feel compelled to add some color. The Tandon program seemed rudderless for a while and got some bad press. Peter Carr was hired to shore it up. He’s done a good job, so far. It’s...

I agree. It just says you've taken a few courses.

B of A has run its investment book there in the past. I'm not sure if they still do.

Me too.

Whomever told you that is an idiot. A big part concerns reg requirements, but there is a huge amount of transactional work as well. Plus the reg work is usually pretty interesting. Desk quants have become a commodity. Having a focus is key.

Lots to address here. 1. Serial correlation, seasonality, mean reversion, time-varying volatility (stochastic vol, ARCH/GARCH). 2. See above. 3. Without liquidity, you have no idea if your price is real. Just because there’s a price observation doesn’t necessarily mean that anything...

I thought you gave that hobby up, Mark.

To do NYU/Courant and NYU/Tandon, I think you'd need the consent of Petter and Peter, respectively. I'm not sure they would grant it.

CCAR 2018 Instructions https://www.federalreserve.gov/newsevents/pressreleases/files/bcreg20180201a2.pdf

@wkdwnstjdTprtm: As usual, @MRoss has hit the nail on the head. Heed his advice.

Plastics

“When in doubt, tell the truth.” I think that this is the kind of thing that could be credibly explained in an application.

Most intro to probability classes focus on the Gaussian and Student's t distributions and only do basic c.i. testing. Did you cover chi-squared and F?Did you compare continuous and discrete distributions? Bayesian inference? Some of this stuff you can teach yourself, but like mathematics in...

Managing people (and quants in particular) is much harder work than you might imagine it is.

I’ve been watching this thread with some interest. This is a topic to which I’ve given some thought. Finance isn’t necessarily a business driven by altruists or idealists. At the end of the day, there’s a profit motive and you don’t find too many folks who are trying to choose between the...

Andy, I took a good look, but the course descriptions are way, way too vague for me to make an assessment. The way the courses are described could easily have been done generically, so it’s hard to see whether they cover academic or practical material. I would need to see a syllabus.

Market risk quants develop and test models, like VaR models and stress test components. They also do model documentation and validation. Nearly all have PhDs or technical masters degrees. BA/BS degrees in anything beyond the governance area is rare. Market risk analysts are usually...

I'm not sure I understand. OR or quant finance? Sure there's an overlap, but are you really doing both? I've met plenty of pde people, physicists, econometricians, mathematicians, and statisticians. I think OR is some of the most interesting stuff on the planet, but aside from portfolio...

Look at the links on the Barra website. I suspect their own research will be of interest.

They do have an advantage. My experience suggests two areas where this is so. First, the students at these generally more aware of what’s happening in the job market for these positions. This is attributable to the fact that many companies have a presence on campus and to the generally...

I have been in the business for over three decades and have worked at several bulge-bracket firms. I have hired dozen of quants - junior and senior. Many have come from "brand-name" schools and have studied math and/or CS. Most have not. I have had many more people come from state colleges...

Change your quantnet username. If people link it to you, it could affect a hiring decision.

6.4. Introduction to Time Series Analysis https://rafalab.github.io/pages/754/section-10.pdf

The most common route to achieve this that I've seen is via hedge funds. A lot of risk people I know have moved to HFs, where risk is often a key part of marketing.

You can build an increase into the outcomes, but it would be deterministic. I'm not sure the covariance structure gives you enough information about directional moves to divine anything useful.

PM me. I can put you in touch with several senior vets on Wall Street who can give you advice. I work with several, including a recovering ROTC Marine vet now very senior in my company.

I've hired dozens of quants. Most are relatively weak communicators. Poor communication skills are probably the single largest impediment that most quants have to overcome. When you're a junior quant at a bank, you work for a quant. When you're a senior quant, you don't. If you're lucky...

PM me. Happy to oblige. As an MS in statistics, I consider myself to be "quant-lite," but I'd be happy to chat.

I did a study for the Financial Services Volunteer Corps a few years ago in Albania one could use other less-liquid benchmarks to price illiquid securities. You'd need access to a Bloomberg. It shapes up to be a panel data problem.

This is true. Some of the best quants I know don't have PhDs. The ex-head quant at Morgan Stanley only had a BS.

Don't write off the possibility of quant finance, however. West Point grads are generally engineers or econ majors, as I recall.

Thank you for your service. I have mentored with a number of vets in this area and know several successful Wall Street quants who also happen to be USMA grads. You should work first, then leverage the GI Bill. PM me.

Always be 100% truthful on your resume.

Net Income After Tax, I think

Well-said. It's ok if you have different stories for different types of jobs describing why you want to do each, as long as the stories are well-researched and sincere.

I'm not entirely convinced that you've thought this through. The business is filled with people who either went into the business on a whim or came in expecting quick riches. Is your interest primarily academic? Have you spoken with people in the business? Do you understand what risk...

Do not reneg on an offer you accept.

Keep in mind that long-term success as a quant is probably more weakly correlated to pure technical competence than you might be assuming it is. The most successful quants I know haven't solved a pde or run a regression for years. They succeed by communicating effectively and by managing...

Rahul, it's very hard to give advice when you specify as broad a spectrum of interests as that. Investment banking, private equity and quant finance are all very different, as are banks and consultancies.

The BIS website (www.bis.org) ia the usual go-to website for the facts. I just found dozens of references. Thereafter, check the Big 4 audit/consulting + McK and OW.

It's not up-to-date, but it covers the bases. Think of it this way: You have two identical swaps, one with US Treasury and the other with a hedge fund that would be rated CCC, if it could get a rating. How do you ascribe relative value?

Start with Canabarro and Duffie. http://www.darrellduffie.com/uploads/surveys/duffiecanabarro2004.pdf

I don't doubt that's true, but I haven't seen all that much of it academically-based. I think it's much more prevalent with respect to shared experience at a particular firm. I can't think of any Harvard cabals in the sales and trading space while I can readily identify the JPM, Bankers Trust...

I have no experience with anyone from Stanford. I have been teaching at NYU for over 10 years. I have been uniformly impressed with the caliber of the students there. I have seen dozens (no hyperbole) become very successful.

1. Good mathematicians do not necessarily make good teachers. 2. Programs generally limit how many courses you can take outside of the department (I have degrees from both and teach at one of these.) 3. I don't agree about the social value and I won't even say anything about the "usurping...

How do you know you want to be a quant?

Recruiters and hiring managers are likely to see them as equivalent.

PM me. Let's chat.

This has not been my experience. Certainly you get the occasional egomaniac that looks down on everyone (they usually don't last) but it's more of a partnership. Consider: the regulatory burden is here to stay. Traders don't have to like it, but they do have to deal with it. Most of those...

Not necessarily a meaningful comparison. The investment bankers often have liberal arts degrees where grades are easier to come by and harder to compare. Lawyers are a mixed bag. Some are 4.0-types. Operations people have varied backgrounds - I have met several who never attendeded college but...

Happy to help. I will be updating this soon.

Start with ARIMA. Then try some kind of GARCH approach. Box and Jenkins are your friends here.

If you're in a good program, I wouldn't care much about your GPA.

Basel III: international regulatory framework for banks http://www.bis.org/publ/bcbs193.pdf

T Be sure to take my risk and regulation classes.

Read the Basel 2.5 and Basel 3 stuff from the BIS. AIRB IDRC IMM CRM

The best of my CGU students are as good as anyone I've seen. I have not met anyone form the Stevens program. It has a great reputation as an engineering school, but my only direct experience there has been as a sub musician in their music department on clarinet.

A couple of things: I'm CRO for the Americas at Barclays - that's plenty. I guess you could describe it as "connecting", but it's simpler than that. If someone : does well in my class (generally a B+ or better) participates (so I get to know them) seems like a fit for an open spot I will...

I disagree. I teach in three MFE programs and I have an MS in stats. There's a huge difference. A statistics MS would likely include: Probability Inference Time Series Regression Multivariate OR Methods Other than time series, I don't think you'd get those in an MFE.

I concur. Contact him. He is a great guy and he is committed to making that program top-notch.

I know plenty of senior quants who have done really well in asset management.

PM me with your email.

Is this for a school paper or for your own trading?

I'm not sure I agree. EURCZK shows clear regime change. MADUSD much more volatile generally, but looks more-or-less stationary. Is it MADUSD you're studying?

This is the best advice you will get. Change the things you can. Get your scores up. Get experience. The only thing I will add is that focus on the industry, not the specific job.

PM me. I know those guys really well.

Your application will be stronger with real work experience. Teach at NYU and Baruch (and at Claremont Grad School as well). I see backgrounds like this all the time. What often distinguished them is the actual work experience.

Weekends don't count unless they are trading days (which they usually aren't.)

PM me. I had had to solve this problem. I know of two different approaches to doing this. One is closed-form, the other is Monte Carlo based.

OR is very mind-expanding and will help you consider alternate ways of solving problems. FFT has some interesting applications in finance (seasonality, etc.), but I'd go with OR.

Work first. Find out what you like. Also, you will be MUCH more motivated to do grad work if you understand why you're studying what you're studying.

Nothing will happen right away. The rulemaking process is slow and cumbersome. All of the required rulemaking from the DF Act isn't even complete. Most likely: the pace of reregulation will slow and banks will have the chance to adjust to it. VaR isn't going away, nor are DFAST or CCAR...

Old guys like myself tend to cut young guys like yourself some slack as long as you don't look too hip or funky. That said, Brooks Brothers is always a safe bet.

Liam, I don't know who you are, but I think Andy should hire you as as an advice columnist. This and your earlier post should be required reading for aspiring quants.

Well said.

1- Do you think that QF was a bubble ? Or the market is soaked with Quants ? If not; What type of jobs still available for Quants QF/MFE ? Yes, - a bit, but there's still real work to be done with real jobs being created in the banking sector. 2- Do you think that DSA is the new MFE ...

1. All quant processes need to be reviewed by quant auditors. 2. & 3. I don't know - I've never worked at one. 4. The engineering program has a great reputation, but I haven't worked with anyone from their MFE program.

From a bank standpoint, quant analytical capacity has become a commodity. We talk about time series people, PDE people, etc. There is still great demand, however, on the risk and audit side. CCAR and other reg requirements are very quant-intensive. There may be continues demand on the HF...

Has anyone tried this?

Dangerous to spline unless you have a very strong prior that that is how the data really behaves. You could Brownian Bridge the quarterly data, but it looks like you don't have too much history so your variance assumption would have to be too strong.

Full disclosure: I'm not a real quant. I have an MS in statistics from NYU, so at best I'm "quant-lite." If you find the markets interesting, it's likely you'll find QF interesting. If you happen to be very quantitative and QF looks like an easy way to make money, you will not like it. I've...

Or Durkheim

Kant - Prolegomena to Any Future Metaphysics

What do you mean by a "quant job?" Developer? Strat? Programmer? Trader?

I'd put them at the bottom where you typically list things like outside interests and computer languages; I wouldn't include them in the "education" section of my resume.

You forgot to mention the cars. Taking care of them can be a real bother. Stop sugarcoating things.

I have worked with him at several firms. You couldn't be more wrong. he's one of the best in the business.

He speaks the truth.

Ken Garbade even gives me a shoutout in the preface - not that that would bias my view.

Fixed Income Analytics

Within the sphere of their commercial interest, yes - firms encourage you to expand your knowledge of the business and usually pay for at least part of any tuition you incur. If you look well beyond your area - if you were a desk analyst who wanted to take law school courses, for example, they...

Not any more. With the advent of Dodd-Frank and Stuructual Reform in the UK, analogies involving gambling houses became passé. Banks are becoming utilities. Underemployed and unemployed traders are more and more commonly found in the banking world. I'm sorry to hear you're bored. I've always...

I was going to respond here, but you've said it better than I could have.

The classic version of this problem (which every finance student in the 90s had to do) is that you are given futures prices for the first n contracts and then are given a series of points further out. From this you have to derive forwards, discount rates, and ultimately yields-to-maturity for...

VaR may not be dead, but it's definitely coughing up blood. It will be supplanted as a reg measure by ES. That said, with banks de-levered and portfolios de-risked, measures of daily trading risk aren't nearly as important as they once were. It's all about stress testing on the banking risk...

If you're not willing to put in this kind of time, this business isn't for you.

Bad move.

Reapply. Be persistent.

Periodically (and regularly, for that matter) big banks simply stop hiring to save on costs, sometimes for months at a time. Usually, this doesn't apply to internships.

Bloomberg has this.

Bloomberg has this daily. The exchanges publish it also.

Year-end hiring freezes are loosening up.

This is all true. Quantquanta: the other posts here are a bit ill-informed. PM me to discuss.

If you're a "back office vp" making that much, I will guarantee you that your life will NOT be stress free.

I disagree. The market risk role could be leveraged into many different areas.

No matter what you do, you will be faced with ethical choices; it's unavoidable. Just do the right thing. I've been in the business for over 30 years and I haven't done anything about which I wouldn't be able to tell my mother.

Or even Rich Uncle Pennybags from Monopoly. He's a hedge fund guy, isn't he?

We might be talking Richie Rich territory here.

I did some work on this. PM me and I will send it. I can also send a sample spreadsheet. I teach part-time and it's a homework I often assign.

Have the statistical properties of financial markets changed since the crisis? Time series: ARCH/GARCH, ARIMA Cointegration Correlation structure PCA of swap curves

It worked for me. Of course, I got my ms (NYU) over 20 years ago.

See what I posted in the books section. In addition, Canabarro recently edited and published a set of important papers in "Counterparty Risk Management", which is published by Risk Books.

Correct. This is frequently used for reg calculations where you need some kind of reasonable proxy.

This is a very slippery slope. The Street is strewn with the careers of people who went into finance for the money. They usually end up stuck in middle management and miserable. Better try an internship and see if you really like it.

I have used GARCH to generate missing implied vol data. Other times series techniques are widely used in reg calculations. Op risk often uses ARIMAX. Incremental Default Risk often uses ARIMA. Monte Carlo (w time series elements) is used in IMM credit. Gaussian Copulae are used...

Are you asking what is used in the real world?

1. Stat skills are very useful. 2. FRM is useless without experience. 3. Time series, database skills, multivariate, regression. 4. PM me. I have coached many of these. 5. Know the regulatory environment,

Do you have access to a Bloomberg terminal? I did this for the S&P 100 once (many years ago) and once I had the symbols, it was easy.

Even better: Truffle Butter

There are dozens. Maybe hundreds. Many are firm-specific. For example, some firms measure CSPV01 while others use PV1% or PV10% for credit spreads. Some firms normalize that measure at the 5y point to eliminate curve influence. Bloomberg will have a decent, if not comprehensive list. Go to...

This is tantamount to saying that you're interested in either oilfield maintenance, becoming a CPA, or playing oboe in the NY Philharmonic. You're going to have to focus a lot more. Right now CCAR and other "middle office risk management" areas are where the jobs are. This may be the only...

What he said.

I think that would be hard to justify. What data, exactly?

I asked my Claremont grad student Shuwen Liang to translate SR 11-7, the Fed/OCC seminal document on model validation, into Mandarin. I'm told she did a great job. Check it out.

While employers in the US are unable to ask about citizenship in the interview process, they are allowed to ask if a candidate has permission to work in the U.S. Many face visa constraints. I have always found that having citizenship / green card / visa status available very useful.

Consider alternatives to a pure quant finance degree. I did a stat masters many years ago. It took me a year (calc + linear algebra) going part time to prepare.

It was my class that made you a quant, I think.

http://www.bis.org/publ/bcbs189.pdf

Time series. Regulatory models.

Not necessarily, not necessarily and not necessarily.

Which employers? Where? There are descriptions in pretty much every risk textbook, but it sounds to me like you're reading the classifieds from 1995.

See what's available from QRM. They wrote the book on it. Their ALM software is used at every ,an or bank.

This was confirmed with HR. I was surprised, too. Many other firms have 3 month notice periods for MDs only.

In which group?

Managers of interns HATE it when their charges actively look for positions elsewhere.

What are you smoking? YES As a senior risk person, I can assure you that for many if not most fields in finance, statistics is more useful than much of what you'd get from a generic quant finance program.

This isn't true. I know plenty of people with degrees from state schools (and by "state" I don't mean Michigan, UVA or UCLA) who got great entry level positions at bulge bracket firms. It takes industry knowledge, persistence, networking skills, and a little luck - but then again, "luck" is...

Wow. This escalated quickly. A few points here: Every senior risk manager I know is well-versed in statistics - and I know a few. The suggestion that we are there to deceive regulators is simply offensive. I'm glad whoever wrote that doesn't work for me. VaR is a relatively minor part of the...

In model validation you are likely to see many different kinds of models, including pricing models, risk models , revenue models, and stress models. There is a lot of demand for model validators.

Figlewski's "Options, Futures and Other Derivatives" providers a superb explanation of option pricing in intuitive terms.

No. Not at all.

Taking this approach at a big investment bank will certainly get you fired.

This is good stuff, Peter. It confirms what I have observed anecdotally. Keep it coming.

You had better be sure about your decision. Reneging on an offer is serious business and is likely to make company 1 very angry. These things have a way of catching up with you...

Credit modeling. IRB stuff, usually leveraging copulae. Frequency and severity of default. Big into tail risk stuff. Also Potential Future Exposure modeling. IMM for derivatives exposure. Deposit/ALM modeling. Heave use of QRM and related tools. Balance sheet management.

Under normal market circumstances, traditional measures work very well. In some markets (eg commodities) they never stopped working, even during the crisis. All models are bad. Some are useful. Many work really well - until they don't. It's easy to be an iconoclast when you're not proposing a...

I worked with the central bank of a Balkan country on a similar issue. We had to price bonds that traded several times per month. I recommended a panel data approach. PM me if you want to know more.

Yes, next weekend -but only briefly.

I am staying in Noida sector 93a. Maybe Monday evening?

I leave in a few hours

If you're worried about having family time, don't go to Wall Street. Things have gone well for me, but I didn't get to see my kids grow up. There are too many quants and not enough dedicated physicians. Go to med school. Trade on the side.

It might be because of the peculiarities of the vol smile in FX. Often there are two different models used - one to price and one to hedge.

I just posted a very interesting 2013 Brookings paper on the Chinese financial system in the risk doc collection. It's pretty short and very well-done. I'd be interested in hearing what our Chinese members think of it. Is it accurate?

Here's a very good piece from Brookings on the Chinese Financial System. http://www.brookings.edu/~/media/research/files/papers/2013/07/01-chinese-financial-system-elliott-yan/chinese-financial-system-elliott-yan.pdf

Well said @mhy. I did an MS is stat/OR 20 years ago at NYU and it was brutal. And I don't think that program was as tough as some of the quant programs out there now. The upside of a part-time program, of course, is a) you save a lot of money, b) you don't lose the work experience and...

The reason is clear: we teach them stuff that is useful.

Statistics will lend more flexibility.

The smile calibration issue with BS for FX is well-known and probably well documented. Some firms have used one model for pricing and another for hedging/risk.

More importantly, I have generally been very impressed with the Cornell folk I've met.

If I were you, I'd go to Cornell.

There are two basic quant career paths, the PDE route and the econometrics route. You seem to be a better fit for the latter, obviously. Many of them are to be found in risk. The challenge for you won't be the math. It will be the institutional complexity. It's not uncommon for firms to...

@IntoDarkness , not everyone who reads this blog will get a big job at BlackRock or Goldman. There is an oversupply of quants. One place where they are still in demand is in audit. When you say "dead end", what do you mean? No job satisfaction? Low pay? No advancement? I know a lot of audit...

I have seen people move between the two groups very regularly. Also, this kind of position could lead to a consulting job, since model review is a key part of CCAR.

It looks like I'll be visiting Dehli in April. I think meeting some students from quant programs there might be fun. Does anyone have suggestions?

I think you'll find they are equally passionate about what they do. If they weren't, they'd find it difficult to teach for two hours on a day where they've already put in 11 or 12 hours working.

Do a Q-Q or a KS.

BUT being a good model validator will give you better job security. The number of models requiring formal review has grown exponentially.

I've seen more people move from liquidity-based jobs to the front office than I have seen move from model validation. (Come to think of it, I can't name a single person who has made that move.) Regarding the regulation comment: get used to it. It's here to stay and will be just as prevalent...

Disagree. Trading is the hardest work you'll ever do and very few do it successfully. On an EV basis, you're much better off in a permanent risk job.

http://acidcow.com/pics/67731-these-gifs-will-help-you-with-your-math-skills.html

I love cookies.

Here's a VaR lecture I gave at MIT. I only recently found out that they taped it. http://ocw.mit.edu/courses/mathematics/18-s096-topics-in-mathematics-with-applications-in-finance-fall-2013/video-lectures/lecture-7-value-at-risk-var-models/

Read their study on corporate bankruptcy. They update it every year. They show one year and cumulative probabilities of default across all corporate issuance by issuance cohort. Do some reading on credit scorecards. Understand how a Merton model of credit default works. Understand...

I disagree. While most quants can code, I wouldn'let a quant anywhere near my production code.

@Osogordo , I've seen career-switchers have varying degrees of success. The key thing is commitment. When I interview people looking to make a big career change, I always ask why. Is their interest genuine or purely money-driven? You need a good story.