EFTA02588164.pdf

From: Vincenzo lozzo Sent: Sunday, August 17, 2014 1:41 PM To: Jeffrey E. Subject: Re: Taxes I guess the bigger question becomes: c=n the parallel between the two fields bring some insight? Because if I'm ju=t head butting against every trading desk in the world and 50years of resea=ch I'm not going to win :) So for instance monteca=lo simulations in my world is called 'fuzzing' - which essentially means th=t given an application you try to feed it with random input to see if it br=aks/crashes Two known problems of fuzzing are:&nbs=; 1) the bigger/more dimensions the input space of the applicatio= the less likely it is to find bugs 2) almost always fuzzin= ends up being a local search - so you need N fuzzers to really try and scr=tch the input space Those two reasons btw ar= why good human bug-finders are a scarce and precious resource <=iv> I guess the parallel there is tail risk - Monte Carlo sim=lations work well assuming you get the distribution right and that you stic= to stuff that is say less than say 6sigma away So=two interesting questions become: 1) can you improve on that? In m= world genetic algorithms are very useful for that for instance <=r> 2) if you can't - can you trigger events that are past t=e frontier of what montecarlo simulations can properly assess? If the answe= is yes and the cost of doing that is limited than you can make money. = Yeah I'll buy to book, I still have to finish the=money biography and Hull's book on derivatives though Sent=from my Iphone On 17/ago/2014, at 09:09, "jeff=ey E." <[email protected]</=» wrote: <mailto:[email protected]> r=sk tolerance, can be accomplished by derivatices. , arbitraged for mi= pricing, risk is never really reduced without a corresponding reward=adjustment, howevr, tax allows certain arbitrage, for ex &=bsp; gains at 20 -losses at 30 Percent . review monte car=o simulations, need resource , bank, friction charges. read fooled by=randomness. On Sun, A=g 17, 2014 at 9:02 AM, Vincenzo lozzo w=ote: Can you let me know if y=u are? I'd be helpful if we can meet up - I'm trying to look at derivatives=through my lens. EFTA_R1_01769023 EFTA02588164  What is somewhat striking is that it seems like all they=are is a programming language whose goal is to minimize risks (with the sid= effect of creating arbitrage and speculation opportunities). What I am is if you interpret trading as an optimization=problem of: given a belief(stock X will outperform the market, etc etc) max=mize returns and minimize risks. Then all derivatives are is 'functions' to=go from a risky bet to a less risky one - now you can go to riskier ones bu= that's an 'anomaly' Anyhow if this parallel somehow holds there are interest=ng questions that come up, for example: what is a Turing machine in this wo=ld? Which would be a fancy way to answer the questions: what derivatives ar= missing? Also since you can stack up multiple 'functions' (eg: combine a f=rward with an interest float-to-fixed swap to make a riskless arbitrage) th=t leads to complexity and hence to bugs But I'm not sure if this is me going insane/trying to fo=cefully fit one thing into the other or if there's actually something there=/div> One of the reason why meeting up would be useful :=) Sent from my Iphone On 17/ago/2014, at=08:47, "Jeffrey E." <[email protected] <mailto:[email protected]» wrote: [email protected]&g=; wrote: taxes are variable friciton On Sat, Aug 16, 2014 at 4:36 PM, Vincenzo lozzo < <mailto:[email protected]» wrote: So quick question for you: are taxes a trick?<=r> Here's what I mean, I think (gut feeling - no proof) that there are potentia=ly a number of arbitrage/risk-free opportunities that are not tackled by an=body because when you add in taxes you don't actually make money. Is 'hacking' taxes a strategy? It seems like all the academic texts leave th=t part for the reader which makes me think there's something there. Also I was reading a while ago that rentech was using a weird structure with=Credit Suisse (iirc) where they were taxed on their profits as if it w=s long-term capital gains instead of short-term. Sent from my (phone 2 EFTA_R1_01769024 EFTA02588165  &=bsp; please note The information cont=ined in this communication is confidential, may be attorney-client privi=eged, may constitute inside information, and is intended only for the use of the addressee. It is the property of JEE Unauthorized use, d=sclosure or copying of this communication or any part thereof is strictl= prohibited and may be unlawful. If you have received this communicat=on in error, please notify us immediately by return e-mail or by e-mail to [email protected] <mailto:[email protected]> , and destroy this communication an= all copies thereof, including all attachments. copyright -all rights re=erved please note =div>The information contained in this communication is confidential, may=be attorney-client privileged, may constitute inside information, and is intended only for the use of the ad=ressee. It is the property of JEE Unauthorized use, disclosure or cop=ing of this communication or any part thereof is strictly prohibited = and may be unlawful. If you have received this communication in error, pl=ase notify us immediately by return e-mail or by e-mail to [email protected], and=br> destroy this communication and all copies thereof, including all attachme=ts. copyright -all rights reserved please note =div>The information contained in this communication is confidential, may=be attorney-client privileged, may constitute inside information, and is intended only for the use of the ad=ressee. It is the property of 3 EFTA_R1_01769025 EFTA02588166  JEE Unauthorized use, disclosure or cop=ing of this communication or any part thereof is strictly prohibited =and may be unlawful. If you have received this communication in error, pl=ase notify us immediately by return e-mail or by e-mail to [email protected], and=br> destroy this communication and all copies thereof, including all attachme=ts. copyright -all rights reserved 4 EFTA_R1_01769026 EFTA02588167