http://www.slate.com/id/2215503/#return Maybe I'm too dimwitted to work there, but I think these questions, or at least the explanations, are wrong. Take the first one, 10 bidding 90, 100 offering 91, what's next price? Anyone who has traded knows it won't be less than 90, but they say the right answer is mid to low 80's. Maybe if you alter the question slightly, otherwise, no way. The coin flip thing, same deal. The theoretical value calculation is straighforward, but only a fool or someone working at AIG would use it to price a ticket to play the game. The explanation ignored to me the salient point--you are playing the game once, not an infinite number of times. Would you risk your entire net worth on a lottery ticket if the odds were favorable?
Bill Eckhardt makes some great points along those lines, AAA. Like he says, if you had 100 billion dollars other than risking all your money for an edge play where you'd be rewarded with immortality there would be NO suitable edge worth risking the 100b. Even if I said you have a 99% chance of doubling your money that 1% risk of being broke wouldn't be worth it.
Agreed, don't really see how they get the answer to the first question. As to the second question, they do mention that you need to talk about marginal utility etc. IMHO, the notion of utility provides one of the few feasible ways to resolve the various paradoxes.
I understand the point of the question. But it must have been mis-written in the article. As written there is no down side - so I would bet everthing I had... Plus the tarp bailout. it says we "stop" it does not say I lose my bet. that first question is a bit of joke as well. In liquid stocks the trade frequently goes to size during the day the session. So there is a reasonable chance 91 gets lifted. Especially if it has been sitting there a while.
perhaps he was describing how DE Shaw's market making algos work. post a fake bid, and when the customer submits a market order, the bid at 90 will miraculously disappear and the customer get filled at 82? "Broker-dealer activities of the D. E. Shaw group are conducted in the United States through D. E. Shaw Securities, L.L.C. and D. E. Shaw Valence, L.L.C., which are registered as broker-dealers with the United States Securities and Exchange Commission and are members of the National Association of Securities Dealers, Inc. D. E. Shaw Securities, L.L.C., a member of the D. E. Shaw group, acts as placement agent on behalf of certain affiliated investment vehicles of the D. E. Shaw group. " http://www.wilmott.com/messageview.cfm?catid=5&threadid=69168
I agree, the first one is simply wrong. The correct answer to the first question is "it's impossible to tell on the given information". For example what if the stock is halted and a takeover bid is announced? The next trade could be 150. The question as stated only says there are 100 people offering at 91 and 10 bidding at 90 - that in no way implies ANY SUPPLY WHATSOEVER at 90.99 or lower. You cannot even claim the price will trade at 90.99 next, let alone right through the bid and much lower. Also there seems to be an assumption that the quantity of buyers vs sellers is relevant. This is not true, it is the quantity of demand and the quantity of supply, not the number of people. Otherwise 1,000 1 lotters would drive the price up when Soros is selling 20,000 contracts. You could even answer "somewhere between 90.32 and 90.33", then when the interviewer tries to contradict you, say "I am 90.32 bid and 90.33 offered". In the second case, again there is no "correct answer" and no wrong answer, since the interviewer does not know your motivations or risk preference.
I think the first question is like soft market. If you want to sell, than you have to compete with other sellers for the buyer. Like a buyers market.
"The type of questions most interviewers askâand those D.E. Shaw is known forâare those with no right answers." The key, from an interviewer's viewpoint, is to observe how you think and reason your way through a problem. If some of the astute points that were brought up here, were made in the interview, I doubt that not coming to identical conclusions as mentioned in the article would be grounds for automatic dismissal. A more black and white conundrum is the monte hall problem; very confusing to grasp the first time around, but rock solid in objectiveness of the answer.