Slashdot Mirror
My Life as a Quant
A complete understanding of Derman's work as physicist, or as finance theoretician, is of course beyond the scope of a memoir. This reviewer studied quantum mechanics in college and took an MBA at UCLA (more about this later) -- adding to my interest in the memoir's technical discussion -- but Derman reasonably pitches his discussion toward a lay audience with many helpful visuals to describe less obvious mathematical relationships. Do not let the perceived arcana of Derman's work keep you away from this memoir.
Emanuel Derman came to New York City in 1966 from Cape Town, South Africa. He started a Ph.D in theoretical physics at Columbia, somewhat in awe to be studying among a cluster of Nobel Laureates. As a teenager, Derman had hopes of being another Einstein if he stayed with physics. But as he notes, time decay happens to ambition. Seven years after earning his Ph.D, he was happy to be an employed postdoc, sharecropping his knowledge of particle physics to willing bidders.
The job market for theoretical physicists continued south. Family responsibilities, his wife's career as a biologist, and iffy prospects for a tenured teaching position --these all added up to Derman abandoning his love of physics, and going to work for money at Bell Labs.
There, Derman fell in love with programming (lex and yacc being two favorite tools). During five years, he built compilers and designed a nonprocedural language, HEQS (Hiearchical EQuation Solver), a precursor to Visicalc. But he never quite adjusted to the politics of Bell Labs, and by 1985, Wall Street was beckoning.
Executive recruiters sought out high-value programmers like Derman. He took a position with Goldman, Sachs in the Financial Strategies Group and began modeling options. It was a good fit. He found himself using sophisticated modeling techniques comparable to "doing physics." Moreover, he soon would collaborate with another Goldman, Sachs employee, one of the most influential theoreticians around: Fischer Black, whose Black-Scholes option pricing model (1973) is a benchmark in the field.
But My Life as a Quant is more than technical discussion; it's also a human interest narrative. The chapter "Easy Travel to Other Planets," about Fischer Black, is worth the price of this book. With compassion and honesty, Derman evocatively portrays his genius mentor. Derman shrewdly assesses what the arc of his life has meant. He shares vulnerabilities, decisions made from the weakness of loneliness, for example. Or, in a self-deprecatory vein, faux pas he committed. He's around Nobel Laureates in both physics and economics, and while noting such illustrious company can at times seem self-serving, the overall effect remains an engaging, complex self-portrait.
One idea about the world of quants Derman dispels is that derivative securities are wholly computer-driven. Despite more computing power on Wall Street, Derman asserts human imagination still leads the way. It takes a Fischer Black to intuit the qualitative to set up the quantitative model. Modern computational tools, however, aid the visualization such creative work thrives on.
As an example of the foregoing, and on a personal note, this reviewer remembers derivative security analysis circa 1969. While pursuing an MBA at UCLA, I did grunt work for a private hedge fund, run out of a Westwood apartment. Technology then was a time-sharing computer terminal and a telephone. The fund strategy was to short warrants and go long on the underlying common stock, where arbitraging opportunities were identified, a strategy borrowed from earlier work by Edward Thorp and Sheen Kassouf. My job was simple: I charted historical price data on clear acetate sheets in colored inks for all outstanding warrants against the underlying stock.
I drew hundreds of graphs, assisted in part by an Israeli graduate student (who had fought in the 1967 Six-Day War). I can't recall his name, but remember that when I'd drop by with more price data, ready to take away graphs, he invariably offered toast and coffee. One morning, I brought yet another roll of graphs to the Fund manager's apartment/office. Steve met me outside, saying he'd just got off the phone with Paul Samuelson at MIT, who wanted to know what our graphs looked like. Samuelson had written an article on warrant pricing, Steve added, which was why he was interested in what we turned up. I knew Samuelson as the author of an economics textbook I'd used a few years earlier.
Another morning, when I motorcycled over to drop off charts, Steve again was outside. He said, "Shelton and Markowitz are here." Professor John Shelton had hired me, of course, but I had no idea who Markowitz was -- he evidently did unspecified work with Shelton. Inside, I was quickly introduced to Harry Markowitz, who unrolled my graphs, becoming immediately absorbed. "Let me get a gestalt on this," was all he said. I didn't know then I was in the same room with the inventor of Modern Portfolio Theory. Now I can say he would see something that maybe a Fischer Black, or, these years later, an Emanuel Derman, might see. When he looked up, he said I did good graphs. I never saw him again.
Years later, I felt honored the low-tech grunt work my Israeli colleague and I labored over had interested those two men, Samuelson and Markowitz. They both received the Nobel Laureate in Economics (1970 and 1990, respectively). My point being -- and I'm sure Derman agrees -- it's not great computers that make breakthroughs in the financial theory. It's great imagination plus the tools at hand! (Obviously, though, computers have changed much of the grunt work.)
For me, My Life as a Quant summoned personal memories, but the odyssey of Emanuel Derman from South Africa to Wall Street is a rewarding memoir for anyone with even a casual interest about how the world of finance is being re-imagined. Emanuel Derman didn't really go to Wall Street to get rich. This memoir is a testament to his true passion in life, whether in theoretical physics, in software programming, or in the modeling of derivative securities. He always wanted interesting problems to work on.
You can purchase My Life as a Quant: Reflections on Physics and Finance from bn.com. Slashdot welcomes readers' book reviews -- to see your own review here, read the book review guidelines, then visit the submission page.
Slate has a thing on computer aided trading today as well: http://www.slate.com/id/2112392/
George II -- Spreading Freedom and American values, one bomb at a time.
Here's the standard book on option pricing:
Option Volatility & Pricing: Advanced Trading Strategies and Techniques
+1 Insightful, -1 Troll. What can I say, I'm an Insightful Troll.
This is impossible. First off there is structural error when you attempt to correlate returns-- the returns of a lot of instruments are far from normally distributed, and I have yet to see a factor model that even comes close.
Secondly, and more importantly, you are not receiving a complete picture if you just look at the numbers in the system. As we are speaking about the global economy, the 'entire system' you mention includes the actions of every single person on the planet as well as the weather, etc.
The quants aren't usually trying to predict the overall movement of the market. This is called a "Global Macro" strategy, and relies mostly on qualitative assessments. Quants mainly work on pricing inefficiencies (arbitrage), which can get extremely complicated. Check out When Genius Failed for an example of a quant-based strategy. (Financial purists please leave me that simplification).
Please, for the love of God, no more car analogies.
The book recommended by the parent is an excellent practical discussion about exchange-traded options. It is not, however, "a good technical introduction to the techniques used for quantative analysis on Wall Street", being neither technical nor about derivative securities in general.
The standard recommendation is Options, Futures, and Other Derivatives by John Hull, and this is in fact a very good book and simultaneously a good introduction to many OTC derivatives. I like Paul Wilmott's book On Quantitative Finance. (This book comes in several versions, some longer and some shorter.)
The books mentioned above stress PDE-based analysis. If you would prefer an approach based on martingale theory, try Financial Calculus by Baxter and Rennie. An Introduction to the Mathematics of Financial Derivatives by Neftci is a more elementary version; think of it as "Stochastic Calculus for Dummies." Neither of these two books contains much information about traded contracts.
"The good reader is a rarer swan than the good writer."
I worked in the field for ten years CBOE / BOT.
Hull's book is where its at. Don't understand it? Then don't jump in.
For the beginner, who's just getting a handle on volatility, etc, I think Natenberg is a far better starting place (just the first 6 chapters) -- everything is explained conceptually. After that, it makes sense to jump into the equations in Hull, etc. Taleb is good once you get the hang of things, too...
there is no thing
what else could you want?
Having a MS in FE I can say that if you can't handle Hull
straight out of the box you should persue other career
objectives.
Also as a former currency and bond trader I can say one of
the issues with modelling in general is liquidity is not
adequately accounted for. It's wonderful to have a
theoretical price, but if the spread is wide enough to drive
a truck through that takes a way a lot of its good.
Likewise when it comes to determining fair market when the
shit is hitting the fan. 1994 was a *very* good year to
illustrate that.
That isn't at all to say modelling and the rest of hte
work of quants is not useful or necessary, just that some
people tend to elevate it to levels beyond reasonable and
worse, apply theory in a vacuum.
But any of you at all interested in this stuff really need
to have a sound grounding in calc and differential equations
at a minimum. A few courses in numerical methods are
helpful too.
HSQuote is a front-end to Yahoo's historic data. Free full-featured demo for 15 days or so. Basically it lets you download years of each-day-end information (open value, close, high, low, volume). In a few miniutes I was able to get years (I set begin year to 1900) of results from the fortune 500 companies (had to find that list separately - then process in groups smaller than 125 tickers). I was all ready to code up some predictive functions to figure out what the market was probably going to do next, if it was a good time to sell or buy or hold, similar to Timing Cube. Oh well, maybe one day.
I am currently reading this book and i am not quite so thrilled by it as the reviewer. My complaints so far include
- i am almost half way through and he has not started working as a quant yet
- Can be boring at times. At one point he starts discussing which radio station he was listening to on his commute. It is inconsistent in content, sometimes very interesting and some time really boring. Maybe it was padded to fill up the required pages.
- Not technical, some of his physics research sounds really interesting, but he does not go into details.
- Not the most lively writting style.
I have no regrets about reading this book, and i will finish it, but i am starting to loose interest in the middle. Hopefully it picks up bit in the second half.
Easy peezy:
http://finance.yahoo.com/q/hp?s=MSFT
There are modules in CPAN to do the scraping for you (check out BeanCounter.pm)
- The Bond and Money Markets: strategy, trading, analysis by Moorad Choudhry
- Liar's Poker by Michael Lewis
- When Genius Failed by Roger Lowenstein
- FIASCO by Frank Partnoy
- Bombardiers by Po Bronson (fiction, but absolutely hilarious and well worth a read)
The last one's fiction, but well worth reading.Obviously, these are all about the fixed income markets, as opposed to equities.
Anyway, having said all that, you can read all the books you want, but the best way of learning the business is to sit on a trading floor, next to the traders.
Jack
Another excellent book that touches on what quants do is When Genius Failed, by Roger Lowenstein. This book charts the rise & fall of Long-Term Capital Management, a hedge fund that relied heavily on mathematical models to guide their trading activity. It's a cautionary tale about placing too much faith in mathematical models of markets that are not always rational.
You want to download data from Yahoo, go to the "historical prices" page for the stock, and look near the bottom: they have a link to download the data in CSV format. You don't need any specialised software for that!
Example: go to IBM's historical prices page, and note the link at the bottom, "Download to Spreadsheet".