how do quants contribute at the trading desk? do these desk quants work with a calculator doing brunt calculations all day?

788 posts, and you ask this question?????It's pretty simple - a good quant will learn more about the weaknesses in his/her model by sitting at the desk and seeing the way the traders use (and abuse) the model. A good trader will be willing to trade new models quickly IF they have seen inside the black-box and understand the way all the assumptions work together. This confidence only comes from seeing the day-to-day changes in the model, hence a close seating position is vital.In short - they are a team, and all teams work better if they sit in close proximity to each other. It allows a great trader to learn to become a good quant, and a great quant to learn to become a good trader.EQ

I was going to reply when I saw this last night, but figured I would leave it a day to think about the wording. EnergyQuant has it exactly right. Quants and techs should be as close to the traders as possible. It enables them to learn from each other and work as a team. The only thing to add to the reply is that a really good trader will be as interested in his quant/tech people learning market skills as they are themselves.

In my experience it teaches quants to respect 2nd order volatility and traders to respect discipline. Perhaps this is controversial...

QuoteOriginally posted by: BetaExoticBetsIn my experience it teaches quants to respect 2nd order volatility and traders to respect discipline. Perhaps this is controversial...What's second order volatility ? .... do you refer to vol-of-vol(or gamma ! -- actually i don't sit on the desk and haven't developed a respect for gamma yet. Don't realize what kind of upheavals it can make. ..... To me gamma has been clumsy greek: so tiny an approximation that I feel like ignoring it out of laziness.When it's so hard to make the portfolio even delta-neutral .... for there are too many ifs and buts to whether we got the delta rights -- since delta computation incorporates the inaccuracies of underying stock model used ... flat vol or even local vol. ... then was the calibration correct.If the calibration is not that great .... what kind of confidence could have in such remote computation as second order greeks).ohh... seems I digressed completely form the topic. Excuse - and if possible give some answer.

QuoteOriginally posted by: EnergyQuant788 posts, and you ask this question?????It's pretty simple - a good quant will learn more about the weaknesses in his/her model by sitting at the desk and seeing the way the traders use (and abuse) the model. A good trader will be willing to trade new models quickly IF they have seen inside the black-box and understand the way all the assumptions work together. This confidence only comes from seeing the day-to-day changes in the model, hence a close seating position is vital.In short - they are a team, and all teams work better if they sit in close proximity to each other. It allows a great trader to learn to become a good quant, and a great quant to learn to become a good trader.EQA few years ago there were traders, but now most trading is done with computers. With traders and their desks gone, quants are stuck in offices all by themselves.

N - "most trading is done with computers". Volume of trades - yes, I would possibly agree, due to the rise in algorithmic trading and more efficient order matching.The idea that there are less traders today than there were in the late 90's - sorry, I disagree. The markets seem bigger and more important than ever. They also seem to be more about traders and quants than traders and salesmen. That's just my perception though, and I'm too tired to try and pull together any numbers to justify it.EQPS BLANTANT HEDGE : Obviously I'm pretty biased by my market speciallity - while energy has boomed from where it was 7 years ago, the much larger equities and FI markets may well have gone through a thinning down process.