the postion I am applying for is Quantitative Research Associate, here's the job description,--------------The Quantitative Research Associate will work on the Index Arbitrage desk. The primary responsibilities will include co-developing statistical analyses and programming.Key QualificationsExpertise in signal processing and filtering techniquesExperience with multi-dimensional quadrate and minimization in computer simulationsExperience in GUI development and real-time data acquisition is requiredPh.D. in Physics, Electrical Engineering, Mathematics, or related fieldA self-starter who thrives in a dynamic, fast-paced environment, and can interact with all levels of positions and communicate effectively--------------I'm a Physics PhD, don't have much finance knowledge. I don't know how technical (finance, math, computer) a phone interview can get. with only a few days left, what kind of things should I spend most of the time on? I am not sure what the Index Arbitrage desk is actually. the firm is SUSQUEHANNA international group, has anybody had experience with them before?Thanks a lot.

Susquehanna is well known and (so far as I know) well reputed.Index arbitrage is the attempt to exploit mispricings in stock index futures. I can replicate a forward contract by {borrowing money, using the money to buy all the shares in the index, receiving any dividends on the shares} and then selling the shares on expiry date of the forward contract. This means that the forward price must be F=S.exp((r-d).t) if you assume that dividends are paid at a continuous rate d. If the forward price is higher than this then (in principle) I can earn a riskless profit by buying all the shares and selling the forward, then unwinding at (or close to) expiry date when the mispricing has gone away.Futures contracts are pretty much the same things as forward contracts other than (1) they trade on exchange and (2) they require variation margin to be posted daily, i.e. each day the position value is reset to zero by a cash payment. This means that the value of a futures contract should include a term for correlation between returns on the index and returns on margin (i.e. interest rate). This effect is very small for short dated futures and can pretty much be ignored.The reason for all the optimisation stuff in the job spec is that, with big indices, such as S&P500, etc, it's often more efficient to construct a tracking basket of, say, 30 shares rather than buy all 500 shares. The trade-of is between transaction costs (including slippage on execution) and tracking error.

If you want to cram, then Options, Futures and Other Derivatives (Prentice Hall Finance Series) by Hull is your best bet, and as such books go, not very expensive.With it you can pick up some jargon, and perhaps more importantly get an overview of how these things are thought about.This reads like an entry level position, so you should promote yourself as a smart newbie, rather than a failed Quant.I have found it useful when doing interviews near the edge of my reach is to simply declare ignorance on certian things, whenthat line of questioning starts.Firstly, it stops them thinking of you as a bulshitter, but more importantly, it leaves more time for them to explore the stuff you're good at.Phone interviews vary hugely in how detailed they get. They tend to be be more fact based "what is an X", sort of thing.50% of all C++ interviews ask you the difference between public private and protected, for instance. (learn this now).My tip in phone interviews comes from bog standard exams. You need to "show your working". So when asked what properties an X has, say "it is a sort of Y, so I'd expect it to have a,b,c," Once you're saying that of course, you may realise that the whole point of an X is not to have property c, so you can say that. Even if you're wrong, and in a real time conversation you will say something stupid, you have at least shown that you know about Y, and have the abiity to reason about it.I have at interview said truly dumb things, however a couple of times I've caught myself, and corrected myself. This has not always failed me. The message is to deal with mis-statements as if you'd written them, press the delete button, and move on, do not let yourself get flustered. One can get to the most powerful job in the world regardless of such errors.A nearly correct rule is that a position goes to the person who comes across as least stupid. It is a game, you will drop points, and score others, whoever gets the most points gets hired.Finally, you've a physics PhD, you're not dumb, you may get this, you may not, but you will get a job, so relax.

How good is Susquehanna? Do they pay well? You can private message me.

hi Matrixhow did you get an interview with them? i'm in the same boat too (physics phd, etc) and i didn't get any responce from them when i sent my resume to them. if you prefer to contact me by email it's vova_149@hotmail.compeacevova

QuoteOriginally posted by: whatIstheMatrixthe postion I am applying for is Quantitative Research Associate, here's the job description,--------------The Quantitative Research Associate will work on the Index Arbitrage desk. The primary responsibilities will include co-developing statistical analyses and programming.Key QualificationsExpertise in signal processing and filtering techniquesExperience with multi-dimensional quadrate and minimization in computer simulationsExperience in GUI development and real-time data acquisition is requiredPh.D. in Physics, Electrical Engineering, Mathematics, or related fieldA self-starter who thrives in a dynamic, fast-paced environment, and can interact with all levels of positions and communicate effectively--------------Thanks a lot.Since I am a physics ph.d. as you are, and i am looking for the job, so my experience might be relevant. Judging from my experiences with phone interviews, they come into 2 categories: one kind will ask you primarily programming questions. the other will ask you some basic programming concepts (such as "what polymorphism is" or "why do u need to declare destructors as virtual?" etc). and then mostly concentrate one the job related questions such as statistics questions, PDE questions, even quite some brain teasers., as i just had this afternoon (they are usually from those big Investment Banks). The job description you put up here seems to fall into the first category. The knowledge of Finance probably won't be stressed (as I have encountered a couple of times). but just to be prepared, read John Hull's Option book as another person here suggested.As to the c++, why not try Bruce Eeckle's Thinkin' in C++.

Hi Phil,More on the interview questions please !!! What is the question exactly (all : programming, stat, PDE, stoc. calulus)Give me some of the names of the banks you had interview and rank them in terms of difficulty of question asked.If feasible, will appreciate if you can list the answers as well.THank youMoney

QuoteOriginally posted by: MoneyHi Phil,More on the interview questions please !!! What is the question exactly (all : programming, stat, PDE, stoc. calulus)Give me some of the names of the banks you had interview and rank them in terms of difficulty of question asked.MoneyI haven't got to the stage when they ask me questions like "What HJM model is" (I even wonder if they will ever ask me!) I can list some questions that have been asked:programming: 1. Why do we need virtual inheritance?Answer: In multiple inheritance, a virtual inheritance will give you only one copy of base object of the type of the common base class. This will reduce confusions in casting.2. What's the use of pure virtual functions? How should we define them? Do we need to implement them?Answer: The use of pure virtual functions are primarily for design purposes, i.e. the architech lays down a design pattern, a chart of functionality for programmers to implement. To define a pure virtual function, you need to use a declarasion like this " pure void func() =0". By defining a pure virtual function, it renders the whole class abstract.In general we don't need to implement the pure virtual function in the class this virtual fuction is defined. But there is an exception: the destructor. You need to implement it. And all implementation of pure virtual functions have to be done OUTSIDE of the class declarasion. Hence they are in general not inline by default.3. What's the use of initializer? What's the order of initialization?Answer: initilizers are to initialize the class members, such as array, constants, etc. The order of initialization, in general, is the same order these memembers are declared in the class, regardless the order of the initializer.4. why do we need to have the destructor declared virtual?Answer: to ensure destruction of objects in the sense of dynamic binding, i.e. the members of the most derived type are destroyed first by polymorphic calling of the corresponding destructor. But if we don't provide ANY destructors when we define derived classes, the compiler will provid a default one. I will try to remember more and now I am in a hurry to head out. Will come back later with more. Hope these are helpful.Phil

this question might be a little late... So what did they ask you in your phone interview whatIstheMatrix and how did it go.Just interested...Thanks!