NakedPuts Wrote: ------------------------------------------------------- > I *think* the OP is making a joke about the number > of people coming here trying to do the opposite, > as if all that’s needed to switch between > industries is some designation. > > I know a few engineers without actual degrees in > the fields in which they work, but it’s mostly > luck and not a career path I would advise one to > take. i am not making a joke, i am tired of compiling information and making decks that will never be seen by anyone.
I like ohai, he’s smart.
I want to get a job in baseball. I’m thinking I’d like to be a general manager or an announcer, probably the radio play by play guy.
eureka Wrote: ------------------------------------------------------- > How old are you? After an engineer turns 40 they > generally take them out back and shoot them. Wow hahahahahhaa
MITBENGFinance Wrote: ------------------------------------------------------- > eureka Wrote: > -------------------------------------------------- > ----- > > How old are you? After an engineer turns 40 > they > > generally take them out back and shoot them. > > > Wow hahahahahhaa ^ Actually, I believe they just hand them the gun and the engineer gladly performs the ceremony…
yeah, copulas and CDO’s, hell yeah. one difference is, keys, that electrical engineering actually *works*. there’s a general trend away from over-exotic, over-complex, over-engineered financial products. vanilla is good. copulas no good
Mobius Striptease Wrote: ------------------------------------------------------- > yeah, copulas and CDO’s, hell yeah. one difference > is, keys, that electrical engineering actually > *works*. there’s a general trend away from > over-exotic, over-complex, over-engineered > financial products. vanilla is good. copulas no > good My point exactly; wake me up when “financial engineers” are able to the likes of create nanoscale circuits and eventually computers using carbon nanotubes.
Mobius Striptease Wrote: ------------------------------------------------------- > yeah, copulas and CDO’s, hell yeah. one difference > is, keys, that electrical engineering actually > *works*. there’s a general trend away from > over-exotic, over-complex, over-engineered > financial products. vanilla is good. copulas no > good this trend is temporary. financial engineering is actually quite hard because your environment is adapting around you. there is no way to repeat experiments.
This is an interesting conversation, given that most of the main contributors are clearly neither financial engineers nor normal engineers.
i think the majority of the main contributors are engineers.
financial engineering is adapting to the environment, not the other way around. copulas and complex abstract nonsense dies out or goes bankrupt, vanilla and common sense evolves. the trend is not temporary, its a paradigm shift in quant finance
SuperiorReturn Wrote: ------------------------------------------------------- > i think the majority of the main contributors are > engineers. Or EDOs like me.
what is EDO?
SuperiorReturn Wrote: ------------------------------------------------------- > what is EDO? Engineer in Degree Only.
so what does that mean? sorry.
SuperiorReturn Wrote: ------------------------------------------------------- > so what does that mean? > sorry. I got my degree in Engineering but I did not work in the field.
Mobius Striptease Wrote: ------------------------------------------------------- > yeah, copulas and CDO’s, hell yeah. one difference > is, keys, that electrical engineering actually > *works*. there’s a general trend away from > over-exotic, over-complex, over-engineered > financial products. vanilla is good. copulas no > good Copulas will come back; synthetic CDOs serve a purpose for investment portfolio views on credit. They will also come back in different forms - correlation products use poisson processes and copulas; you need these technologies currently to price baskets of underlyings. There are some issues which are cleaning up in the securitization business: 1) The models used are too simplistic, not the other way around. One of the most highly used (abused) CDO pricing mechanisms is the One Factor Gaussian copula which assumes constant correlation between underlyings. The maths are easier and you can pump more of these things out to the market to collect more fees. There exist higher dimension problems by assuming different correlations for the underlying but these are substantially more difficult to price. Additionally, getting away from normal distributions adds additional complexity as well. 2) Bad incentive compensation structures for D-Bag bankers = any garbage gets securitized and stuffed into clients portfolios (fiduciary duty anyone?) 3) No income loan docks and no standards for underwriting = throw keys in mailbox when times get tough.
eureka Wrote: ------------------------------------------------------- > How old are you? After an engineer turns 40 they > generally take them out back and shoot them. This is true, actually. Happened to my father and most of his peers. “Engineering” changes so fast that old timers are usually obsolete before they hit 50.
adding some complexity to copulas by assuming non-gaussian distributions or non-deterministic correlation structure will achieve what precisely? make the “pricing” of these structured products even less transparent and abstract. overloading the model with complex stuff is not an obvious step in the right direction of understanding the risks associated with these products, it’s an attempt to hide the issues in greek formulas. the failure of this approach is imminent. put the elegant and beautiful copulas in academic papers and stats books where they belong for now. anyone who thinks they understand correlated credit default processes is dellusional or academic. plain vanilla is the answer
Mobius Striptease Wrote: ------------------------------------------------------- > adding some complexity to copulas by assuming > non-gaussian distributions or non-deterministic > correlation structure will achieve what precisely? > make the “pricing” of these structured products > even less transparent and abstract. overloading > the model with complex stuff is not an obvious > step in the right direction of understanding the > risks associated with these products, it’s an > attempt to hide the issues in greek formulas. the > failure of this approach is imminent. > > put the elegant and beautiful copulas in academic > papers and stats books where they belong for now. > anyone who thinks they understand correlated > credit default processes is dellusional or > academic. plain vanilla is the answer …where do babies come from?