SERVING THE QUANTITATIVE FINANCE COMMUNITY

 
User avatar
coffeeOneSugar
Topic Author
Posts: 17
Joined: January 29th, 2009, 9:11 am

hitting the books, but what books?

February 21st, 2009, 4:12 pm

I recently graduated in theoretical physics from one of the top two british universities. I have spent the most of the last 4 months calling HHs, sending a billion of CVs, etc. As any self-respecting grad student preparing a reconversion in quant. finance, I carefully read Hull, two Wilmotts, and a couple of other books on quantitative finance, and I now know almost by heart the solution of every brainteaser ever invented since Oedipus met the Sphinx on his way to Thebes.However, the very few interviews I managed to obtain were always for jobs involving black-box/automated trading (I have very strong programming skills, thanks god). I have a good knowledge in statistics, but definitely not of the kind that was required. Especially, I know very little about time series analysis, and I find it very difficult to find good books about it. I understand that most trading algorithms are proprietary, but I was wondering whether there was a couple of *good* books explaining the use of time-series analysis in the context of high-frequency trading. My understanding so far is that time-series are just simplified (linearized) neural-networks.Also, but this is subject to debate, I feel I have been wasting my time studying Hull and Wilmott, because it seems to me that most investors are not going to spend much on complex derivative products for at least a decade, and so IBs are not going to hire entry-level guys like me for a long time from now. I am happy I have learnt a lot about mathematical finance, which is a *mathematically* terribly appealing subject, but I have the feeling that most people now doubt about its practical relevance. So it won't help me find a job. Black-boxes and automated hedging or investment strategies, on the other hand, have proved more solid and sustainable.
 
User avatar
Nomade
Posts: 169
Joined: January 19th, 2008, 10:26 pm

hitting the books, but what books?

February 21st, 2009, 5:14 pm

Ok, I will try to help you. First off, to say time series is like neural networks really shows how bad your understanding of this stuff is If you want to learn about time series, take a look at Robert Almgren NYU Courant home page -- he teaches a course on the subject and post lecture notes online, in addition to other references. Now, as for complex derivatives -- yes, the big clients got burned pretty badly, but there are still some activity going on, although at a fraction of what it used to be. I'd say a decade is perhaps too long. Some areas are actually growing (eg equity derivatives, FX & commodities structuring). The problem though is that for each seat, there are hundreds of suitable candidates + strong incentive to hire internally. The other thing is that Hull is light for someone aiming at being a derivatives quant. There are many books that can be helpful. I would suggest, as a follow up to Hull, that you try Baxter/Rennie and MJoshi's book. Reading the books might not be enough though, so I suggest you read it and get your hands dirty trying to implement some of the models. If you can read French, I also suggest reading El Karoui's notes that is available on her website. As for HH: different strokes for different folks, but in my view it is much better to find a good HH and stick to him than distributing your CV to everyone who posts a (fake) job on willmott.com. In a market like this, personal contact is probably more useful in getting your CV across than a HH. I can refer you to some HHs that I had good experience in the past (PM'me for details). Last but not least: try to talk to people in the area and figure out what you want to do -- that is the most important thing at the entry level. Once you start, it is much trickier to change areas, so be sure you know what you're getting into.
Last edited by Nomade on February 20th, 2009, 11:00 pm, edited 1 time in total.
 
User avatar
GoingQuanting
Posts: 9
Joined: December 20th, 2008, 5:31 pm

hitting the books, but what books?

February 21st, 2009, 11:13 pm

Here is a commonly used text for Financial Time Series analysis but not sure if it will get you all the way to blackbox trading systems, .Analysis of Financial Time Series.
 
User avatar
deepvalue
Posts: 918
Joined: April 25th, 2007, 6:08 am

hitting the books, but what books?

February 22nd, 2009, 6:38 am

QuoteOriginally posted by: coffeeOneSugarBlack-boxes and automated hedging or investment strategies, on the other hand, have proved more solid and sustainable. Why would an investor dump his money into a "black box". By definition, a black box is something he doesn't understand. Why would investor want to dump their money into a black box? Just go play the slot machine!
 
User avatar
TamasRS
Posts: 28
Joined: January 31st, 2009, 5:00 pm

hitting the books, but what books?

February 22nd, 2009, 11:12 am

Well, I have very similar experience. I also have a PhD in physics and I am looking for jobs in this area (I didn't find anything else to do, except game programming, no jobs in academia, no jobs in industry.... ). On my all interviews people told me that the high-frequency trading is the future and almost nobody does quantitative finance anymore. Of course these interviews were with HF hedge funds, and they are selling themselves, but these are the only interviews I got.TRS
 
User avatar
GoingQuanting
Posts: 9
Joined: December 20th, 2008, 5:31 pm

hitting the books, but what books?

February 22nd, 2009, 3:24 pm

QuoteWhy would an investor dump his money into a "black box". By definition, a black box is something he doesn't understand. Why would investor want to dump their money into a black box? Just go play the slot machine! I think its an unfortunate phrase. Maybe, I am wrong, but I just take that to mean some form of automated trading system that doesn't require humans to make the trading decisions.I have no idea what state of the art is in this area but there are some interesting articles on this site on it like Thorpe's on the system he was using.
Last edited by GoingQuanting on February 21st, 2009, 11:00 pm, edited 1 time in total.
 
User avatar
coffeeOneSugar
Topic Author
Posts: 17
Joined: January 29th, 2009, 9:11 am

hitting the books, but what books?

February 22nd, 2009, 6:22 pm

QuoteOriginally posted by: NomadeOk, I will try to help you. First off, to say time series is like neural networks really shows how bad your understanding of this stuff is If you want to learn about time series, take a look at Robert Almgren NYU Courant home page -- he teaches a course on the subject and post lecture notes online, in addition to other references. Thanks for the reference :-) Almgren's notes are brilliant. I am not quite as impressed by El Karoui's, but anyway...
 
User avatar
KackToodles
Posts: 4100
Joined: August 28th, 2005, 10:46 pm

hitting the books, but what books?

February 22nd, 2009, 10:10 pm

QuoteOriginally posted by: TamasRSOf course these interviews were with HF hedge funds, and they are selling themselves, but these are the only interviews I got.TRS HiFreq HFs tend to be more receptive to (some would say desperate for) physics/math/CS types because their style of trading (being essentially technical analysis and high-speed tea leaf reading) is largely discredited by the 'serious" academic quants. Thus, the only phds who are willing to play their game are the influx of physicists/mathematicians who don't know anything about finance. The problem with these high freq funds (aside from their fact that they are tea leaf reading) is that their turnover is lso high frequency. They trade stocks in and out -- just like they trade their employees in and out! The second you don't make money, they replace you. Blink!
Last edited by KackToodles on February 21st, 2009, 11:00 pm, edited 1 time in total.
 
User avatar
KackToodles
Posts: 4100
Joined: August 28th, 2005, 10:46 pm

hitting the books, but what books?

February 22nd, 2009, 10:13 pm

QuoteOriginally posted by: GoingQuantingthere are some interesting articles on this site on it like Thorpe's on the system he was using. How many people can continue to make money using automated systems like Thorpe on a sustainable basis? Alpha is a zero-sum game. If everyone is using an automated system to make money, where is the money coming from? Remember, physicists, money is conserved. Not everyone at the poker table can win every hand.
Last edited by KackToodles on February 22nd, 2009, 11:00 pm, edited 1 time in total.
 
User avatar
phil451
Posts: 83
Joined: December 7th, 2007, 8:21 am

hitting the books, but what books?

February 23rd, 2009, 8:17 am

Hmmm... the totally spurious science of Time Series Analysis. Heavy theory to back up a number of doubtful assumptions!The alchemy of financial markets.Anyway a really good book is 'Time Series Analysis' by Hamilton. The mathematical treatment is excellent and the text is clear and concise.
 
User avatar
coffeeOneSugar
Topic Author
Posts: 17
Joined: January 29th, 2009, 9:11 am

hitting the books, but what books?

February 23rd, 2009, 9:45 am

QuoteHiFreq HFs tend to be more receptive to (some would say desperate for) physics/math/CS types because their style of trading (being essentially technical analysis and high-speed tea leaf reading) is largely discredited by the 'serious" academic quants. What really matters to me is that serious quants are in serious troubles these days, and that the only ones recruiting now are HiFreq HFs.What is a serious quant? One who takes the money in the good days and blame the models for their simplicity in the bad ones? To be honest, mathematical finance is a beautiful subject, but I am not 100% sure that believing that prices behave like pollen floating on water beats tea-leaf reading by much.
 
User avatar
2face
Posts: 3
Joined: June 29th, 2008, 6:51 am

hitting the books, but what books?

February 23rd, 2009, 5:21 pm

I second the book recommended by phil451, 'Time Series Analysis' by Hamilton...
 
User avatar
QuantVader
Posts: 247
Joined: July 10th, 2008, 2:19 pm

hitting the books, but what books?

February 23rd, 2009, 6:55 pm

I don't think that money is conserved. It is called the money multiplier effect.
 
User avatar
KackToodles
Posts: 4100
Joined: August 28th, 2005, 10:46 pm

hitting the books, but what books?

February 24th, 2009, 4:38 am

QuoteOriginally posted by: QuantVaderI don't think that money is conserved. It is called the money multiplier effect.Yeah, just like the subprime investors "multiplied" their money? Or the internet bubble speculators "multiplied" their money? Or Madoff and Enron "multiplied" their money? The "multiplier effect" from monetary flow theory has NOTHING to do with quant investing, friend. In the investing world, multiplication = ponzi scheme. You better learn basic common sense before your quant strategies "multiply" your investing dollars out of existence. There are always losers at the poker table, friend!
Last edited by KackToodles on February 23rd, 2009, 11:00 pm, edited 1 time in total.
 
User avatar
QuantVader
Posts: 247
Joined: July 10th, 2008, 2:19 pm

hitting the books, but what books?

February 24th, 2009, 7:02 am

Your argument would have to come from having a fixed global capital and this is true only for short periods of time. Economic fluctuations change the size of net alpha over time, but having long-term economic (positive) growth creates the possibility of having net positive alpha since we have more wealth and more capital.
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