Search results

  1. Yike Lu

    MFE: To do or not to do?

    I wouldn't use the word "flexible". I think the functional style and the extra math does make functional languages more rigorous, but it removes some flexibility. It's always a tradeoff. Flexibility vs strong guarantees. Python to me represents the higher flexibility end of the continuum. C++...
  2. Yike Lu

    Where should I focus on if I would like to switch to quantitative finance

    Reading the market and economy -- as in a traditional analyst? That's not a match for your skills, but as it with anything it can be done. More likely would be something statistics and machine learning based.
  3. Yike Lu

    Where should I focus on if I would like to switch to quantitative finance

    Not necessary. You should try getting a few interviews and see how you fare, as a first step. No harm, as any "wait X time before re-applying" is going to be less than the time for MFE. Depends on what kind of QF job you're looking for. I'd do some research on the landscape of roles if I were...
  4. Yike Lu

    No Longer Seeking Quant Jobs - Market is Changing

    lulz, so much logical fallacy in this thread.
  5. Yike Lu

    P&L of Delta Hedging strategy

    At each time point, compute the delta hedge. Execute the delta hedge. Mark all delta hedges to the final underlying price. Sum up PNL.
  6. Yike Lu

    MFE: To do or not to do?

    Software engineering skills become even more important in that context, although you can lean on other people to help teach you. Really the important thing when hiring is that the quant recognizes the importance of software engineering as core skill to be practiced and improved upon. The old...
  7. Yike Lu

    MFE: To do or not to do?

    Once again, it's not for day2day dev work, but rather for something like an ORM layer, debugger, IDE, or when you need to add in a truly new feature and can't wait the (2011 - 1998 = 13) years for somebody else to put it in (think auto and anonymous functions). A workaround by definition is...
  8. Yike Lu

    MFE: To do or not to do?

    I don't see how duck typing / dynamic typing is worse in terms of "customization". It's weaker in safety and guarantees for the obvious reasons, but they provide better flexibility (which is our chosen metric here... though there be some disagreement). And for the record, what I mean by...
  9. Yike Lu

    MFE: To do or not to do?

    Well now we get to the heart of the matter. "Can be extended by generic libraries" -- does this not describe every modern language though?
  10. Yike Lu

    MFE: To do or not to do?

    No, you brought it up originally, referencing how customizable C++ was. What does it mean to you?
  11. Yike Lu

    MFE: To do or not to do?

    What exactly does customizing a language even mean to you then?
  12. Yike Lu

    MFE: To do or not to do?

    Ignoring other languages as it's not germane to the discussion. C++11 doesn't have dynamic types, nor automatic memory management. Smart pointers aren't automatic memory management (as attested by many segfaults that can still happen when they're used), and auto is not dynamic typing, it's...
  13. Yike Lu

    MFE: To do or not to do?

    When I asked about customization, I was expecting you to use your particular expertise to filter down the possibilities and give some real examples. It's not helpful to point to a link. Also, when "customization" is mentioned, I mean meta-programming facilities, such as classes that create...
  14. Yike Lu

    MFE: To do or not to do?

    I don't either, but it's the gold standard in terms of customization. You can use the language to change its own syntax. Not having seen the meta-level things the language can do, your picture of programming is fundamentally incomplete. Any concrete examples of customizing C++? It's useless...
  15. Yike Lu

    MFE: To do or not to do?

    Disagree. If you want maximal customizeability, you go to LISP. Python is certainly above C++ for customization. C++ in my experience is rigid, it just covers a ton of cases within that rigidity. If you want to wander outside of that, be ready to get hacky.
  16. Yike Lu

    MFE: To do or not to do?

    C++ is uniquely useful for the subset of problems that require all of the following simultaneously: (1) Fine grained control of multi-threading with shared memory (2) Very low-latency (3) Relatively high level of design abstraction. Remove (2) and you can use Java [and Java is still decent...
  17. Yike Lu

    MFE: To do or not to do?

    I don't believe the path dependency of learning programming languages. It is true that C++ is harder than Python, so more C++ programmers can pick up Python than the other way around, but if one is capable of either, no reason one can't learn Python first.
  18. Yike Lu

    MFE: To do or not to do?

    Whether 'tis nobler in the mind to suffer the slings and arrows of outrageous fortune, or to take up arms against a sea of troubles and by opposing end them.
  19. Yike Lu

    (Master) CSME @ UCSD or Bioinformatics @ SDSU

    Uhh, this is a quantitative finance forum. We know very little about the programs you mentioned.
  20. Yike Lu

    Undergrad education advice B.Com vs B.Sc

    I don't see this is a large barrier. You can probably write "Bachelors" and be fine. There is enough fragmentation in what BS vs BA vs whatever else means in the US and in the world generally that this goes to least common denominator of "Bachelors".
  21. Yike Lu

    Undergrad education advice B.Com vs B.Sc

    This feels like a very Canada specific question btw. Not sure an American hiring manager would even know the difference between BCom and BSc. Higher chance of an American MFE director knowing though, as it's more relevant.
  22. Yike Lu

    PhD in physics, first job as a quant

    It's an optimization problem where you need to balance getting practice with potentially burning opportunities (at a certain time scale, as you can re-apply after a while).
  23. Yike Lu

    So I am 2 months away from getting my MFE. Now what?

    1) Yes. Path of least resistance and least risk. 2) Highly idiosyncratic and dependent on the specific job requirements. 3) Not my area. 4) Leverage first, especially if you're still employed while looking, as you have more optionality. If you were unemployed, there's always a timing problem...
  24. Yike Lu

    "Becoming" a "Quant"

    Python is definitely recommended. Better to go Python then VBA than the other way around, as you develop good habits that way. I don't know your level of computer knowledge, so to cover all bases it's better to go with an online course first, as video lectures and homework therein will cover a...
  25. Yike Lu

    COMPARE UC Berkeley MFE vs NYU maths in finance

    Being in NYC exposes you to far more opportunities at small firms, firms that don't have the time/money to go out to recruit on the West Coast. However, discovering those opportunities relies more on networks (whether you go out and network yourself or you let the school network do the job).
  26. Yike Lu

    Hedge Needs - Which Model is Best to use?

    Option Traders Use (very) Sophisticated Heuristics, Never the Black–Scholes–Merton Formula by Espen Gaarder Haug, Nassim Nicholas Taleb :: SSRN
  27. Yike Lu

    Interviewer was wrong about an interview question - correct him later?

    It's pretty simple, telling your interviewer is a high vol move. They'll love you or hate you. If they already love you, why risk it? On the other hand, if the interview was ehhh or they hated you, there's nothing to lose.
  28. Yike Lu

    Hedge Needs - Which Model is Best to use?

    If you're only trading vanillas and have a decent measure of the distribution, try pricing using the p-measure expectation just by integrating over the probability distribution (this is the important piece, not so much the hedging), and hedging using standard BS. SV models are for pricing...
  29. Yike Lu

    Trading Interest Rate Swaps

    There's been a push to centralize clearing of IRS for exactly those reasons. See Cleared OTC Interest Rate Swaps There are also techniques that account for this type of risk. CVA is one.
  30. Yike Lu

    Know anyone with GRE 160Q got admitted to top MFE programs?

    Why don't you just retake it and not screw up?
  31. Yike Lu

    Quant developer or Software engineer?

    Pure Matlab production code is an anti-pattern. Same could be said of R.
  32. Yike Lu

    Automated trading - room for humanz?

    Predominantly in fundamentals, M&A, macro picture.
  33. Yike Lu

    Automated trading - room for humanz?

    There are pretty well defined boundaries of what machines can or can't do, whether it's because the algorithms aren't developed enough, or because the logistics simply don't allow it. Yes, many manual trading tasks are being automated, but there are niches where that isn't practical.
  34. Yike Lu

    Automated trading - room for humanz?

    Can a machine drive out to a Wal-Mart, measure foot traffic, assess aesthetics/presentation/customer service? And repeat for a well chosen statistically sound sample across the US and abroad?
  35. Yike Lu

    Need advice for en engineer looking to become a quant

    CFA L1 helps you talk the talk. I haven't taken L2 or L3. L2 I've heard is very useful if you want to be in corporate finance / M&A. You'll do fine, best thing is to go on a few interviews first to get calibrated on what's required vs what you have. Don't try for your dream job in this phase...
  36. Yike Lu

    GRE Score. Which to submit?

    Hmm... that's tough. One the one hand, higher Q percentile is what's really important, on the other hand that's a large gap in V percentile.
  37. Yike Lu

    GRE Score. Which to submit?

    What are the percentiles on each of the individual scores?
  38. Yike Lu

    How does quant pitch a stock?

    Given you want to do quant investing, it's unlikely that you will want a job for which "pitch a stock" is an important interview question.
  39. Yike Lu

    What is a trading analyst position like?

    The blue skies pure math research is ever more rare in industry. The areas where that was applicable (exotics, mortgages, credit) are declining due to bank regulation. To be fair, they may arise again in a different form, so long as people want to trade these products. I do a fair bit of...
  40. Yike Lu

    What is a trading analyst position like?

    Speaking strictly to the point of @The Spartan Guy being a PhD in Computer Engineering, it's much more likely that a buy side firm will have better IT environment for you. I've worked on both sides of the fence, and the one time I was at a sell-side shop, the IT situation was massively...
  41. Yike Lu

    US Citizen - Mention on Resume?

    I donno, I'm of the opposite opinion. Nationality <=> Nationalism. Citizenship is a legal status.
  42. Yike Lu

    Optimal portfolio - dubious weights

    I believe this is a commonly known problem of mean-variance optimization -- it has a tendency to go to corner cases. You're going to have to add some type of "diversity" metric or shrinkage to the utility.
  43. Yike Lu

    Getting S7 Sponsorship

    It's very off the beaten track so it's case-by-case. One firm may think it's the stupidest thing ever, another might love it. Sounds reasonable to me, though.
  44. Yike Lu

    Is MFE worth it if I am already working in the industry?

    Not sure name-brand value of MFE will help you. Perhaps specific course work will. If I were you, I'd come up with a proposal specific to modeling/research and ask my current company. You don't need a graduate degree, but you need to have a fairly high level of mastery of a broad skill set...
  45. Yike Lu

    Quant only for PhDs?

    Risk is the major one I can think of, then model QA/validation. Quant trading would be another, though much harder to get into and highly individual whether or not one is suited.
  46. Yike Lu

    Quant only for PhDs?

    There are still firms that only hire PhDs, although this appears to be more for marketing purposes. Also certain roles need to be filled with subject matter experts (ie PhDs) as opposed to generalists (MFEs). However, there is definitely a market for the generalist.
  47. Yike Lu

    What OS does quant use most frequently?

    Manual trading => Windows. Automated Trading => Linux.
  48. Yike Lu

    Purest way to trade option skew?

    So if I'm reading you right... (1) Means you don't want the squared effect? Does this really matter a lot to you? If you can predict future vol, you can predict future var. Also the fair strike var swap won't equal the squared vol swap strike, but again why does this matter? (2) So you REALLY...
  49. Yike Lu

    Purest way to trade option skew?

    Put another way, the replication / hedging strategy is exactly what you need to do to trade skew.
  50. Yike Lu

    Purest way to trade option skew?

    That's a bit short sighted. How is a variance swap constructed? From vanilla options -- the strike price of a var swap reflects market option implied volatility, not time series volatility.
  51. Yike Lu

    Studying FE but Don't Want to be a Trader

    The likely career path would be something like... FE => quant role for a few years to develop product expertise => salesperson or consultant about that particular product. You can go straight into such a role, but I don't see that very often (could be a bias where few FEs want to do it).
  52. Yike Lu

    Purest way to trade option skew?

    Did a quick Google for "skew swap" because I couldn't come up with the answer myself. This Wilmott thread looks promising... Wilmott Forums - Skew swaps
  53. Yike Lu

    I'm not a math genius. Do I have a shot?

    Gotta keep the devil way down in the hole...
  54. Yike Lu

    I'm not a math genius. Do I have a shot?

    I don't know what kind of stereotypical "math genius" you have in your head... You have to be smart enough to understand advanced math and apply it. Which kind of math is highly domain specific... you won't be needing differential geometry or group theory for example. You don't have to be able...
  55. Yike Lu

    Wasted time, bad decisions, bad GPA

    Why haven't you tried just applying to places or getting in touch with a recruiter to do so? This can be for internships. See what's out there and see what you're missing skill-wise before deciding to commit to an Masters. You simply don't have enough information to evaluate what how employable...
  56. Yike Lu

    Current PhD. Student Considering the Switch to Quant.

    String theory is not even close to being testable. At this point it might as well be religion. The LHC is 27 km in circumference. To reach the energy levels required to test String theory with current understanding, I remember reading the track would have to be as large as the solar system.
  57. Yike Lu

    How do bank trading desks actually work?

    Check with the banks. I don't know if that's a commonly traded product. If not be ready to expect high t costs.
  58. Yike Lu

    Current PhD. Student Considering the Switch to Quant.

    I don't at all consider it a shame specifically for Particle Phys (HEP + theoretical), since the applications of the actual discoveries will be nil for many years (IF they arise at all). The sideshow technical innovations (HPC and ML specifically) are where you see people branching off into...
  59. Yike Lu

    Current PhD. Student Considering the Switch to Quant.

    Specifically for High Energy... (1) It sounds cool, discovering the mysteries of the universe. (2) It ends up being a lot of grunt work for little pay and the mysteries of the universe won't have any real-world application for probably 100 years.
  60. Yike Lu

    Current PhD. Student Considering the Switch to Quant.

    Those are good basics. I would add the following things (generally in the CS department, not C++ specifically): * General comfort with Linux command line tools * Comfort in Excel with VBA * Basic functional programming. * Python. * Software engineering / design. The last is kind of hard to get...
  61. Yike Lu

    Request for Application Evaluation

    Based on this: https://www.ets.org/s/gre/pdf/concordance_information.pdf, you should definitely retake. In the old GRE regime, 800Q was the standard. Look at how quickly the percentile falls off...
  62. Yike Lu

    Modeling Forward Contracts, Option Prices, Spot Prices Together

    Stochastic forward curve with stochastic volatility.
  63. Yike Lu

    Career switch for PhD: skills question

    No, you don't need industrial grade C++. However, if you do not start second guessing your error checking and corner cases once you hear those magical words "this is going into production", it's a major red flag.
  64. Yike Lu

    Current PhD. Student Considering the Switch to Quant.

    Hi Robert, everybody says they have good C++ and data skills, but I have found in interviewing people that the variance in what that means concretely is actually quite high. As such my best advice would be the following: (1) Get the lay of the land here by reading some of the job guides...
  65. Yike Lu

    Quant Straight Out of Undergraduate

    Yes it's possible. UofT is a disadvantage relative to the Ivys, as many firms simplify by hiring exclusively from top schools, but not insurmountable.
  66. Yike Lu

    Uniqueness of Ticker Symbols

    I'm not able to directly answer your questions, but I know that in practice, equity shops have to maintain their own unique security IDs internally and that this is a non-trivial task which includes looking over Corporate Actions (CAX) daily.
  67. Yike Lu

    Personal Advice regarding MS.

    MFE is designed to get you a job in (most likely since they have the most jobs) an investment bank. For what you want to do (Central Banking), Math Econ seems the most appropriate fit. Math Stat is a bit too general. However, I don't know the quality of the programs between MFE vs Math Econ vs...
  68. Yike Lu

    Laptop Thread 2015-2016

    I'm no longer a student. I like my MacBook Air, but I recommend getting your hands on Win7 + Office somehow, for Excel, as a practical matter, as it's widely used. Fortunately, Windows is unnecessary at my current job.
  69. Yike Lu

    Personal Advice regarding MS.

    GRE should be fine. The math section is ~1st-2nd year university math, no advanced stuff. Verbal is tough, but not super relevant for MFE.
  70. Yike Lu

    Laptop Thread 2015-2016

    The reason to use Windows is not VC++, but Excel. I personally avoid doing real work on Windows.
  71. Yike Lu

    Macro Level Questions

    You're competing against seasoned professionals who have more experience, knowledge, time, technology, data, manpower, and economics (better brokerage / exchange fees etc). As such, I don't think it's a wise use of your time, long term. By all means check out some of the free data sources out...
  72. Yike Lu

    Texas Quants?

    I'm from TX, not living there anymore. Back in '09-'10 when I was at home looking for quant jobs, most of the buzz was around Houston for energy trading. I know one prop firm located in ATX, and there's a decent hedge fund scene in Dallas, although the ones I was looking at aren't quant.
  73. Yike Lu

    Please provide advice for a prospective quant

    There are certain things in CS that make you a better coder but have very little direct application. OS, compilers, functional programming each have limited direct day to day application -- unlikely you'll need to write an OS, a compiler, or to use Haskell -- but make you a better coder via the...
  74. Yike Lu

    Please provide advice for a prospective quant

    Learn how to code. Learn statistics and preferably machine learning. Learn networking and social skills. Everything else is domain specific knowledge.
  75. Yike Lu

    New here and in need of starting tips

    So coding is a weakness, but besides that I don't see why you believe you need something extra special to break in. Have you tried yet?
  76. Yike Lu

    New here and in need of starting tips

    You seem to have the background required, in which case you'd just apply to jobs or contact recruiters. Not sure what you're worried about. What are your weaknesses? How's your coding?
  77. Yike Lu

    Will I have time to play video games?

    Forget the US scene, South Korea is where it's at... http://www.nytimes.com/2014/10/20/technology/league-of-legends-south-korea-epicenter-esports.html Where men with names like Na_lra, Boxer, Nada, Savior, and Bisu walk as legends. Where 300 APM is passe.
  78. Yike Lu

    Self employment on resume

    My rule of thumb is that you have to be ready to answer any questions raised by a seeming inconsistency or conflict of interest. If you aren't then avoid it.
  79. Yike Lu

    Coding academies as gateways to $100,000 jobs

    They're selling the shovels. To be fair, shovellers themselves are in high demand.
  80. Yike Lu

    Proving your Programming Skills

    Being from TX, nobody ever calls it U of T, just UT, or if you're from OK, you call it TU.
  81. Yike Lu

    Proving your Programming Skills

    U of T could also be Toronto, and possibly Tennessee.
  82. Yike Lu

    Functional Programming

    It's not a matter of workaround, but rather the point of Monads. Using tuple is an implementation detail. Call f and h functions with type signature (double -> double). You have demonstrated how to compose f and h. From what I understand, this is a specific subset of the application of Monads...
  83. Yike Lu

    Functional Programming

    This isn't really the canonical application of Monads, although valid. A practical example would be to pass a C style success/failure code alongside the result when each function only takes one arg (x) but outputs two (x, failure code). One thing I've personally done is use the State Monad to...
  84. Yike Lu

    Functional Programming

    I think I've seen the Milewski future talk as well.
  85. Yike Lu

    Functional Programming

    I remember watching this: https://channel9.msdn.com/Shows/Going+Deep/Brian-Beckman-Dont-fear-the-Monads And reading this http://blog.sigfpe.com/2006/08/you-could-have-invented-monads-and.html
  86. Yike Lu

    Pros and Cons of Certain Quantitative Languages

    I don't write it, it's just Numpy is a wrapper for Fortran BLAS libraries.
  87. Yike Lu

    Pros and Cons of Certain Quantitative Languages

    Really the combination of languages is you want one compiled heavy duty language and one higher level interpreted / research language. The former is meant to be industrial grade, building things that last and need to be stable. The latter are meant for rapid development. Examples: Compiled: C...
  88. Yike Lu

    Pros and Cons of Certain Quantitative Languages

    Yeah, there are various ways of going about this. Many things can interface with both C++ and databases. I tend to go Python only these days, but a lot of the heavy duty math that my Python code does is ultimately in C or Fortran (conveniently wrapped for me by someone else). I rarely code C++...
  89. Yike Lu

    Pros and Cons of Certain Quantitative Languages

    Yeah, I've done various things with combining C/C++ and Python. At a very basic level, I've used Python as a high level wrapper to interface with a database and feed input parameters into a C++ function. It's also possible to directly inherit C++ objects as Python objects using boost-python...
  90. Yike Lu

    U.S. Schools Expelled 8,000 Chinese Students

    I think we have the makings of QuantNet's first race war.
  91. Yike Lu

    Pros and Cons of Certain Quantitative Languages

    Python and C++. Python for research, soft realtime apps, and getting things done in general. C++ for the most performance sensitive applications. R is not a good language to build infrastructure in, even statistical infrastructure. The foundation is crooked.
  92. Yike Lu

    Quant Trader: Nature or Nurture?

    The trader mindset: entrepreneurial, gets things done, likes to take risks while also respecting them, has an opinion and willing to back it up with money, conversely discounts opinions with no money riding on them, detail oriented especially with regard to possible tail risks.
  93. Yike Lu

    Worst and best fonts to use in resume

    Yeah, it's not really too relevant. I try to use something with efficient tight spacing so things fit onto one page. It's funny that this showed up in BBG, the person they interviewed is definitely on the artsy side of things; most of us wouldn't even be able to tell the difference except maybe...
  94. Yike Lu

    Worst and best fonts to use in resume

    $ grep -i font ~/.vimrc set guifont=Anonymous\ Pro\ 12' set guifont=Terminus\ 9
  95. Yike Lu

    Business major transition to quant?

    Yeah, you can't just be after the money. But at the same time, you can't dismiss the money either. At the end of the day, your Sharpe ratio is a pretty good measuring stick of how good you are, assuming you are on the trading side of quant. More generally, money is a (very noisy) measure of...
  96. Yike Lu

    Any advice for a 19 year old undergrad?

    That's not true. Risk optimization (in the Kelly sense) grows the whole pie. But agree that a lot of the bad has come from over emphasizing zero sum behaviors and ignoring traditional manufacturing.
  97. Yike Lu

    Business major transition to quant?

    The main thing you're behind on is math skills. The typical applicant will have studied Linear Algebra, ODE, PDE at least. Programming is a nice and practical thing to learn, but math is the largest reason you haven't gotten into quant programs. As for how to remedy, I don't really know, others...
  98. Yike Lu

    Any advice for a 19 year old undergrad?

    That kicked off quite an interesting discussion. "You know what the trouble is, Brucey? We used to make shit in this country, build shit. Now we just put our hand in the next guy's pocket." -- from The Wire
  99. Yike Lu

    Any advice for a 19 year old undergrad?

    Probably because it's all ancient and Chinese like, so it sounds cooler to the kids.
  100. Yike Lu

    Looking for advice

    Yeah, there are pretty well established paths to doing business modelling and research, the classic equity analyst role. CFA will help for sure. Quant is much more specialized, more hard math involved. If you don't enjoy probability and stochastic calculus (among other things), you're not cut...
  101. Yike Lu

    Any advice for a 19 year old undergrad?

    Art of War has been a finance cliche ever since Gordon Gekko recommended it to Bud Fox.
  102. Yike Lu

    Looking for advice

    I don't have any useful alternatives because it's not clear what you actually want to do.
  103. Yike Lu

    A realistic look at the MFE

    This is true. If you wanted to do pricing model research you should have done a PhD. It's that simple. But I don't think most people here want to do pricing model research, and they don't consider that to be the definition of a quant job.
  104. Yike Lu

    Any advice for a 19 year old undergrad?

    Python. Learn it. I'm convinced that the first circle of Hell is a million LOC MATLAB codebase. The second circle is Java, the third is C++. Of course, sometimes you have to go through Hell to get where you want to go.
  105. Yike Lu

    Looking for advice

    I typically don't like to discourage people, but I have to agree. If being a quant is your end goal, then you've seriously mis-invested your time and money, not once but twice.
  106. Yike Lu

    Career Prospects (Silicon Valley VS Wall Street)

    For specifically IBD work, check out http://www.mergersandinquisitions.com/ The Cliff notes of that is it's not quant. There are specialty quant jobs in these divisions, but traditional "investment banking" ain't it. If you don't want to work at a bank, your options are basically hedge fund or...
  107. Yike Lu

    Zero% Strike Floors

    You can't use a shifted SABR? Or maybe a shifted SVI?
  108. Yike Lu

    Zero% Strike Floors

    In a GBM, a zero strike call is the same thing as futures contract. If you want to assume that negative rates are possible, then clearly GBM is not the right model, you'll have to do ABM and adjust surface models to match, which unfortunately means you have to re-derive a lot of it on your own...
  109. Yike Lu

    7.7 Jane Street coming to campus!

    The "Estimathon" is their substitute for mock trading. Trading is basically making an estimate, establishing confidence intervals, then converting those things to a 2 way price, and controlling the risk thereof.
  110. Yike Lu

    Pricing Options Close to Maturity

    Check out the work of Alireza Javaheri.
  111. Yike Lu

    Possible to re-enter trading with bad P/L after additional education?

    You are operating with insufficient information. Get an internship and see what it's like and you probably won't be asking these questions. You might even ask your co-workers at your internship..
  112. Yike Lu

    Optimize reinvestment by flattening

    http://en.wikipedia.org/wiki/Kelly_criterion
  113. Yike Lu

    FO Internship vs BO Full-time position

    I believe the general sentiment is that trading is more difficult to break into and offers more options coming out, regardless of whether one does it successfully or whether one wants to do it longer term.
  114. Yike Lu

    Brokerage connection

    Your broker must offer an API. If they don't, you'll have to find one that does or do some kind of hack around driving a browser or other front end using code.
  115. Yike Lu

    Last quarter of univ. Help me pick classes.

    The point of taking a compilers course is almost never so that you know how to build a compiler. Ditto for s/compiler/os
  116. Yike Lu

    Rotman international trading competition 2015

    Hah! Never thought Rotman would make me famous. Edit: Woops only 164 views.
  117. Yike Lu

    Derivatives trader looking to jump to academia/quant PM role

    The extra stuff you want to learn is along the lines of applied math like statistics, optimization, software engineering, ie the data science toolkit. But it's not very useful without trading / finance / quant knowledge. The stuff that isn't so useful from an MFE (to me at least) is the exotic...
  118. Yike Lu

    Derivatives trader looking to jump to academia/quant PM role

    (a) and (b) could not be more different. (a) does exist. It's hard to point to a specific degree that gets you everything you need, as none of them in my experience do. The closest for (a) is probably MFE and for (b) stats, but there's going to be both (1) a significant amount of extra self...
  119. Yike Lu

    London!

    And Eastcheap was... cheap.
  120. Yike Lu

    Why MBAs get all the gold??

    Put it this way -- most quant jobs are in banks, and banks are about doing business. And doing business means talking to clients, managing relationships, and getting deals done. ABC, always be closing. Quants are a necessity for various reasons, but by and large they aren't the revenue creators...
  121. Yike Lu

    Reduced Course Load?

    It's more about what you can do as opposed to how quickly you crammed in coursework. That said, a good MFE is fairly intense, so one has to consider whether one would be able to handle it.
  122. Yike Lu

    Derivatives Quant or Algo Trading?

    That sounds pretty accurate. Only thing I'm not sure about is exit opportunities.
  123. Yike Lu

    I need a plan.

    Or maybe even ODEs in high school.
  124. Yike Lu

    UCB MFE Computer Science Minor Worth It?

    Depends entirely on the job. Mine is heavy Python and some C++ and a tiny bit of C#. On balance, you're probably right that more jobs are going to involve VBA and Java. That said, if you're a good programmer, you should be able to pick up a new language quickly. I was able to make production...
  125. Yike Lu

    Calibration on heston Stochastic Volatility Model

    Ask your school to see if they have access to any option databases. Yes it is much harder to come by than data for underlying, there's simply much more of it. If your school doesn't have it, you might need to make a connection with somebody at a financial institution.
  126. Yike Lu

    UCB MFE Computer Science Minor Worth It?

    You realize that virtually everything a quant does is programming based, right?
  127. Yike Lu

    Calibration on heston Stochastic Volatility Model

    Good option data is hard to come by. Ask your school.
  128. Yike Lu

    Quant Dev to PM, options and how long

    I don't understand why you looked outside. You are optimally positioned to launch a few strategies, you just have to convince your boss. All your ducks are basically lined up. Worst case at your current firm is you get too small a %, but the upshot is you start getting a track record.
  129. Yike Lu

    CV Latex Template Advice

    US resumes are usually 1 page no photo. Honestly, I tried the LaTeX approach before, it's really cumbersome for a resume. You're trying to squeeze maximum use out of a limited space, which means you want to really fine grained control of everything, the antithesis of how normal LaTeX works...
  130. Yike Lu

    Heston's stochastic volatility model with jumps

    This might be what you're looking for. http://www.math.ku.dk/~rolf/bakshi.pdf
  131. Yike Lu

    Derivative payoff question

    Wow that turned depressing quite quickly. Vaulting ambition that o'erleaps itself indeed.
  132. Yike Lu

    Derivative payoff question

    In the secret parts of Fortune? Oh most true she is a strumpet.
  133. Yike Lu

    Derivative payoff question

    There is a tide in the affairs of men, Which taken at the flood leads on to fortune, Omitted all the voyage of their life Is bound in shallows and in miseries. Brutus
  134. Yike Lu

    Derivative payoff question

    I think this may be more appropriate. What a piece of work is a man! How noble in reason, How infinite in faculty! In form and moving how express and admirable! In action how like an Angel! In apprehension, how like a God!
  135. Yike Lu

    Quant Programmes with solid bases in Behavioural Finance

    I see -- it's a splitting hairs factor to let you decide between top programs. Again, I don't think any QF program will have a focus, there may be one or two courses, but beyond that you'd have to check the school itself.
  136. Yike Lu

    Quant Programmes with solid bases in Behavioural Finance

    Behavioral finance is a sideline for quant programs. Your utility function for quant programs needs some serious tuning if that is a big factor in your decision.
  137. Yike Lu

    High Frequency data download

    20 years might be too long. For academic projects, you usually go WRDS; your school should have account. https://wrds-web.wharton.upenn.edu/wrds/
  138. Yike Lu

    3rd year mid-sized HF quant vs junior BB structurer (seek multi-asset PM long term)

    You want to run a quant HF strategy that trades exotics? You don't really need structuring / BB quant experience to effectively trade vanillas.
  139. Yike Lu

    New book "Advanced Quantitative Finance with C++"

    heh heh heh. @Daniel Duffy knows what I'm on about.
  140. Yike Lu

    New book "Advanced Quantitative Finance with C++"

    Where do you work again?
  141. Yike Lu

    Should a sophisticated model like local vol be for pricing futures and vanillas?

    Let me make this really simple. Buy this book and read it. http://www.amazon.com/The-Volatility-Surface-Practitioners-Guide/dp/0471792519
  142. Yike Lu

    Career transition from Software Programmer to Quantitative Analyst

    MFE is much different than MF. Also MFE is what you should do.
  143. Yike Lu

    how to get stared?

    Well if you're using TDAmeritrade, you should look up their API (hopefully they have one) and start getting data from them. Also Google and Yahoo Finance have APIs. The rest is kind of up to you to apply your knowledge.
  144. Yike Lu

    how to get stared?

    C++ is good to learn to get a job and for low-latency trading. If you're not latency sensitive, might as well learn Python and R. You'll iterate much faster, plus the upgrade path in terms of speed is actually not bad in Python (ie Cython).
  145. Yike Lu

    Determining if FGLS is a better estimator than OLS

    There should be a test for heteroskedasticity? And the sum squared errors should still be valid.
  146. Yike Lu

    Do you even lift? What are your hobbies outside of the quant world?

    I don't really periodize, I prefer evidence based progression. For a lot of my assistance work for example, I use Charles Staley's EDT method. Set a timer (good values are usually 10 minutes or 15 minutes), pick a set of exercises (can be just 1), pick a weight, do as many reps as you can in...
  147. Yike Lu

    Do you even lift? What are your hobbies outside of the quant world?

    Huh? They're competition maxes, ie 1RM. Yeah, it is a relatively weak bench. Relatively long arms I guess.
  148. Yike Lu

    Do you even lift? What are your hobbies outside of the quant world?

    I'll look at that, could be interesting to do some data science on my logs. My experience with app logs is they're usually not flexible enough for what I'm doing. Maybe I should write my own.
  149. Yike Lu

    Do you even lift? What are your hobbies outside of the quant world?

    @Ddavis, that was all raw with belt (USAPL NJ States). Most of the people at the USAPL meet were in the Raw Division, and anyways USAPL Gear Division is 1-ply only. @Anik Roy, the minute you buy a belt you'll hit 3 plate squat. I don't see the point in waiting. Also Oly shoes are debatable for...
  150. Yike Lu

    Do you even lift? What are your hobbies outside of the quant world?

    Thanks -- that's not bad, but if you get a serious belt, you should easily add 50 lbs to that squat. DL improvements vary. I don't run a specific program, and what I do run is highly variable. I try to keep some consistent elements -- squat and bench ME every week, deadlift ideally once a week...
  151. Yike Lu

    Do you even lift? What are your hobbies outside of the quant world?

    429/249/501 SQ/BP/DL -- competition maxes (* in LBs). Done @ weight of 185 (198 class).
  152. Yike Lu

    Someone please explain backwardation (as seen in wiki)

    Also, strictly speaking, "backwardation" is futures below spot. "Contango" is futures above spot. "Full carry" is futures at the cash and carry arbitrage price. Futures should almost never trade above "full carry" because of the arbitrage, so long as there is sufficient storage.
  153. Yike Lu

    Someone please explain backwardation (as seen in wiki)

    This arbitrage is not realizable because you cannot short the spot commodity. Yes, buying the futures/forward is cheaper than storing and carrying, but the inventory risk is high and the benefit is limited to those who actually use the commodity so not as much capital can flow into the trade...
  154. Yike Lu

    A high-frequency trading model with IB

    Like the use of live data. Dislike the haphazard half live/half backtest way you do it. This code does not deal with any of the truly difficult issues of an intraday strategy like limit order placement and multithreading. Look at the IB Java API for hints on how your architecture should look.
  155. Yike Lu

    Large Number of International Students in MFE Programs

    This is correct, however there are a few nuances. First C++ is preferred and proficiency in it is more portable (C++ signals you can learn Java and C#, the opposite isn't necessarily true). I personally never "understood" why because pointers were always very simple to grasp for me, but here's...
  156. Yike Lu

    Large Number of International Students in MFE Programs

    "Best" is probably the wrong word. It's more like some combination of competence and having enough resources (HR, capital). A 3 man hedge fund is much less likely to want to go through the headache for example.
  157. Yike Lu

    Undergrad preparation to becoming a quantitative analyst

    Make connections, learn about the field, polish your presentation skills, go to the career fair. What you're doing course wise seems like it'll be good for the resume. Ask some people you know for some examples of professional resumes, maybe get a book about resume and interview prep. Some of...
  158. Yike Lu

    Undergrad preparation to becoming a quantitative analyst

    PhD is nice, but that decision is very far down the line and very much depends on your next few years in university. If I were you, I'd try to do a tech internship, a finance internship, and a summer research project in the field you're considering a PhD in. It will give you a flavor of what...
  159. Yike Lu

    Georgia Tech to offer an online Computer Science master degree

    Yeah it could be beneficial. I don't know about employer point of view. My opinion is that programming is absolutely necessary for quant work, so if you suck at it, you will hate your life as a quant.
  160. Yike Lu

    Georgia Tech to offer an online Computer Science master degree

    I'm not sure why I'm even defending Java. My first language was Scheme; I learned Java on the job in essentially a week.
  161. Yike Lu

    Georgia Tech to offer an online Computer Science master degree

    How much difference is there really, from a practical perspective? GC (ie no pointers) and a kludgy "everything must be an object" philosophy (which makes things harder, not easier), no OS/architecture specializations required. Not being rhetorical here. Now presumably, there's C in the OS...
  162. Yike Lu

    Georgia Tech to offer an online Computer Science master degree

    Modern tech startups more and more de emphasize low level optimization (read C++) in favor of rapid prototyping and iteration (read dynamic interpreted languages -- Python, JS, Ruby). So I don't see "Lot of Java and a wee bit of C++" as a bad thing, or as programming/engineering skills getting...
  163. Yike Lu

    Getting to the cutting edge

    This is not recent, but it's a timeless classic: http://www.lucent.com/bstj/vol35-1956/articles/bstj35-4-917.pdf Generally speaking, you want to look at things in Machine Learning and statistics as a bsae.
  164. Yike Lu

    The Fear Index

    We'll just have to agree to disagree.
  165. Yike Lu

    The Fear Index

    I got that. I prefer eg something like this http://blogoscoped.com/archive/2007-09-17-n72.html
  166. Yike Lu

    The Fear Index

    Looks awful. Sentient AI jumping out of trading algos is just... really implausible. If I had to gamble on such an event, I would place my money on facebook or Google.
  167. Yike Lu

    Seeking advice regarding buying a new workstation

    I would recommend not using SQL Server. There are several possibilities for alternatives, ranging from NoSQL (Mongo, Redis) to traditional OSS SQL (Postgres) to ad hoc (serial C structs read using Python glue). The main things you'd want in a server are memory (16 GB+) and fast network...
  168. Yike Lu

    Need help choosing a major (Highschool senior)

    You guys must be trying to flatter me with references to Caltech.
  169. Yike Lu

    Need help choosing a major (Highschool senior)

    I looked at the Math degree requirements. It's a generalist math degree, so you can probably customize yours to be "applied". I said applied because there sometimes is a sharp divide between pure and applied math majors, depending on the school. There really isn't an ideal second major or even...
  170. Yike Lu

    Need help choosing a major (Highschool senior)

    Physics, applied math, or econ. Also, IntoDarkness is right. You may be able to do something with Portland, but TBH you're going to have to hustle, network, etc.
  171. Yike Lu

    You need to be able to write custom code to handle big data processing. There may be pre-built...

    You need to be able to write custom code to handle big data processing. There may be pre-built applications, but those usually cost money and aren't portable, or flexible. You should look now. Take a few interviews and see what people ask you; adapt your studying to what they ask.
  172. Yike Lu

    You should be able to answer algorithms and data structures questions. You can showcase work by...

    You should be able to answer algorithms and data structures questions. You can showcase work by working on open source projects and posting the work in github. You need to have experience working with big data sets (>100MB, pref >10 GB). The physics degree says you're smart, the CS stuff says...
  173. Yike Lu

    Depends on what skills you already have. I can't tell just from knowing you have BS Physics. It...

    Depends on what skills you already have. I can't tell just from knowing you have BS Physics. It depends on your programming ability more than anything else. You just cannot do any meaningful work with data these days without serious programming. Actually, I work in a Chicago market maker now.
  174. Yike Lu

    What is the job satisfaction of financial engineers?

    MATLAB probably not. C++ most likely. Python or any of the web languages (hint hint) most definitely.
  175. Yike Lu

    Quant Jobs for Masters in Physics

    No, not detrimental. You know there are a LOT of physics people in FE / QF, right? I'm a BS in Physics, and there are plenty of PhDs.
  176. Yike Lu

    Quant Jobs for Masters in Physics

    It's going to depend on what you want to do (Wilmott career guide is helpful). Obviously probability, statistics, stochastic calculus, option pricing basics, programming (these are the common core competencies). The general trend I see is towards machine learning (which extends beyond finance)...
  177. Yike Lu

    Can an OLD man get in here ?

    A place that executes orders for customers, without taking proprietary risk. Usually they are specialized to take large blocks and move them over the course of a day or more. Of course, more quant knowledge would help, but seems like you have the software engineering chops (especially...
  178. Yike Lu

    How to proceed with coding this trading application

    Java will be easier. For frameworks, look for concurrency frameworks and GUI frameworks. Java has standard libraries for both. NB. easier != better
  179. Yike Lu

    Can an OLD man get in here ?

    You cannot get an algo dev job as is? If not at a prop firm, I would think agency execution shops could use your skills. People have gotten such jobs with far less CS than you have already.
  180. Yike Lu

    Python relevance

    "Use the instance in the application." As for how to create the class: http://henry.precheur.org/python/Dynamically%20create%20a%20type%20with%20Python Map the XML to that by walking the parse tree. More specifics and I'd be writing the app...
  181. Yike Lu

    Python relevance

    I just think it's very easy to find out the answers yourself. Also, you want a yes/no or an actual discussion of what's normally done? 1. Yes, pretty easily. 2. Yes, but you generally don't need to "compile" to DLL, because it's a dynamic language. You can directly use the new class at runtime...
  182. Yike Lu

    Python relevance

    Dynamically creating a class is really standard (and eminently Google-able). The DLL part is usually not done, and kind of not the right mindset. You COULD compile it down to C dynamically and load that, depending on the type of code, see for example Numba.
  183. Yike Lu

    Python relevance

    Yes, maybe you've heard of it: http://lmgtfy.com/?q=python+generate+code+from+xml This isn't exactly a highly unusual problem, so your first expectation for Python really should be that it's solved already. I can actually see where you're coming from though, there are many things I would...
  184. Yike Lu

    Python relevance

    Umm, there's a library for practically every standard thing in Python, so yes. There is in fact an XML parsing library (several in fact). And with the right imports, it's also quite easy to parse the mathematical expressions. The only thing I don't quite understand is why you persist in being...
  185. Yike Lu

    Python relevance

    Yes, but it's still not as intuitive. data.table is VERY clean and always gives me what I expect. I can do very quick ad hoc data exploration with it. Part of it is that I haven't fully learned pandas, but that's also partially pandas' fault. Data.table maps so well to SQL concepts: dt[where...
  186. Yike Lu

    Python relevance

    Python >> (Matlab, R) when it comes to software engineering tradition. Can't speak in detail about Matlab, but R is a hodgepodge of packages that are rarely designed with inter-operability in mind. I still like it better for EDA because of the data.table package, but for any non ad-hoc data...
  187. Yike Lu

    Hedge Needs - Which Model is Best to use?

    You're already fighting a losing battle trying to hedge an exotic without recourse to vanillas. You cannot hedge factors that aren't traded.
  188. Yike Lu

    The steampunk future

    His schtick about how technology could have been different is certainly interesting. I like it, if only for the fact that so few people try to think that way. But the idea that a slide rule can in any way come close to competing with computers is so laughable that I wonder if this guy has any...
  189. Yike Lu

    R language

    It's used, although I venture to say that Python is taking over its place in many ways.
  190. Yike Lu

    Necessary to learn C#?

    Thanks for asking a question then proceeding to completely ignore the answer.
  191. Yike Lu

    Necessary to learn C#?

    It's good for everything except low level, low latency, and explicit multithreading.
  192. Yike Lu

    MSc. Theoretical Physics or Computer Science

    What subject did you do your undergrad in? Apparently not physics... Leaving aside the "inner workings of the universe" thing (which btw I've been through already), what "opportunities" are you talking about? Especially for theoretical physics. I see your job path out of it as being limited to...
  193. Yike Lu

    Necessary to learn C#?

    The way I see it breaking down, the startup world is dominated by FOSS languages, especially dynamic weakly typed ones like Python, Ruby, Javascript. C# has the unique (dis)advantage of being the dominant Windows language. On the plus side, there are lots of people developing for Windows. On...
  194. Yike Lu

    Question on an exercice of Joshi's Book

    The formula for \(S_T\) has nothing to do with the payoff function at all.
  195. Yike Lu

    Princeton MFin Still Any Chance For Princeton Interview?

    I interviewed with them a few years back (2010). They emphasized in that interview that "We train leaders, not quants."
  196. Yike Lu

    Listening to music while studying??

    RIP Jaco. He went before his time.
  197. Yike Lu

    Mathematical Models for Sports Betting

    I've taken a look at this, but not a serious one. The thing is, like with any source of data, it costs money to get a good clean source of data. If you were to try to extract it from play-by-plays, you would spend 80+% of your time on text processing heuristics to try to figure out what happened...
Back
Top