Matematica finanziaria 2

Binomial modelthe )(" ()( have)callEuropean Asymmetricoption toStccs.tl payK St youoKmox-= - -: - ,?fairthe trivalueWhat timeatis ↳ without arbitrage opportunity

)Forward ( boreconstruct don't anythingFCT.tl to haysymmetric contactSt K you: - -- asset classfDerivatives classifications : payofffunctional shape( )MODELToy BERNOULLI t ( )MODEL t freeriskyimterest rate- I rateo =,We twoconsider assets bond)I B } ( )B BCI=L ): : Ito R=2) stock S 1"s ": , see ,go fs= =differentTwoto drtimes d 1sIs ... , { ( value t)Vt StBt int StI--of bondsnumber{ × - ( )- ValuePortfolio :( imN o"I "B" "" " "+"x. ×: = "number riskyof assety . ). yep{ (value)x.VI z in( ,It sR y z .t= .. ,↳ variablerandomportfolio (Def )Pluto ) machinewithstrategy Vcd and =LARBITRAGE makingMoneyo-: = - (Proposition de theto dominatethe allowedthe ofreturn stock not returnArsinoeneo isonit n e: u the bond and versaviceProof : he ( 4) foliofront07=0V( o xYs ysts⇐ xx = == - ,. { }fhe slateVH1 by))I )od assumptionSuypose ZItr Ys t Iu tx s t ssite = so= =- -, .}({ )hiVIL ) l ) let ) bySuppose BailItr Ysd t s -2txrit = assumptionSZR> = su so, =, - .To holds thatarbitrage ofequivalentthethat it combinationtocondition is ith is convex usaysay amo . ,dand i.e :.,{ dIt gov OldR t= psohohlities!Hur t theof': Bernoulli;D u "Yg %!: -'; %= = ufor G-hnefolili C④d)im which ④be Qinterpreted )=qties prob and tmeasureoldqu oldsonewascan a i =.. ., ( ) calledSCI ) QE )@dgu RISKin ProOld MEASUREss t NEUTRALtr BA= u s Bluth. .. = -.Eafscsl] discounted)Gtr expectationSco )s- .- .( ]/ XmFmE Xm =L MARTINGALEI. - behaves martingalediscountedThe asset priceBtoN aas. . !binomialWe extend modeltowant to BINOMIAL TREE MODEL3LsdSu/sur I/Sdf this allfree thearbitrage mln tag modelhorriblemodel issuedsu ⇐I -/( Sud)Sco s= freearbitrage'd aresu"Sd 'sLet's Sumodeltoythe fconsider applied to Cloclaim ) with )contingent Scot Slo- ssa - =.: - ,s !shouldthetwqnonille equalbeoutput(that tohas haveportfolio theconstructSuppose risklessto it evolutionwant isawe same)the bondof portfolioat the riskeenisspin .f ] )(Clu Dsd) (d) because(d) it's)ItuSsu lessriskCIT ss ITc == -=-, -- (d))Tcu ITwhere) assetolddu the) portfolioof riskyb is quantity lessobtains rinkim order- a= d)(s u -Moreover - take) TWthi --T ( )Etat4%19,24%11,2)GtrITCH tht doitcast su s= -. alsothediscountedyEQ4d) )" I fun" " d)'t " )Old behaves%-do ) Lstm In call price, t seco u =-= =( )dS u ,- It R !martingaleaasEMCH ) I 9,4dByfsn)In questdo) .= -binomialthe haremodelFor stepwith we3 :( ]Ccl3D" µ÷np HttKY (KY( ttsquqd (tf E a)98 't ( ')Clo olds Sd'3%2 suedsus Sud t%-= --- -,@ ( ]IEQ ¥4 (F) tDNKstern 9T "guk "( )- -CCN) q¥µnBinomial su k)CcoWITHMODEL N STEPS -= =i ( -, ( callofpayoff theflutie )to T!Consider derivativegenerala s: µ ,Approach1) DELTA HEDGINGi )I alreadyfindto computed)It portfoliothat riskthemakes lemwant swessit .. ,ApproachE) REPLICATIONi Inportfolio theconstructwant shouldto betht) arbitragetothis pricesTIN ordersit ovoidwe a -. - .the same : "" "I {fin:'Ii: ':::: ' www.noron-r.no=. . "j""ofpayoff the derivativederivativep ,Vote Ho) Hal to= martingaleItequivalentbondof measuretheq et o ]utld "flu told) )) lolold I)- "E9,101dL-IHol sten ×e thIn -guta= s --- t= -Tf. -.ULt dR s-fo2% iwhere d. u. -"t paranoiano{it,I .STEPTIME .iDef ARBITRAGE: )Vimeo( PC )VHno , A Istep, me 00so, 00m , =Def (financing selfSelf Thewithportfoliostrategy ) tconsider xt.ttteea o a- =: --. ....,financing )Icondition is Xt YtYt StXt t t Sttr t-: - + e., DIt (portfolioeach Atholdoftimethat thet the int hwuho.seequals theYetvalue createdat marketsays ., ., )Gets (of Mets)the heldValue which formed ttsuntilportfolio andtatisnew INTEGRALSSTOCHASTIC )( proofnoLet to's the timecontinuousfrom now . . .walkConsider randoma : l÷i÷÷÷÷÷÷÷÷:¥H Xltof )variableBernoulli Yi Bl ) independentYiYiis Iktaxsequence =a : n .,.IE Hit{ EIHHI #= EEHM.io.][ Yi fE ISx Dx 0. -= - -- (? )@xY.ztC-sxT.f.sxzVAnCxCt5t-iEyVAnC4iT_mCs.xWhat YiVANabout the variance =I' ')Csx) = St 1¥ [ ]firm' 'tif ) with 'St @ order Htt oand o vanso sox some -- =- Ix d lso-st will linearlyso timeincrease in- >This called theis DIFFUSION OF RANDOMBINOMIALLIMIT WALKDef )BROWNIAN ( WIENER ProcessMOTION:) WG)I so ( ) independent(Process tWetwith Wuindependent2) increments WrVVS etand rrr usare: - -( hstationary welProcess with only3) increments dehemdswt om: -a+.Nfo ) theWen to) t continuesvarianceh , f"¥%:%"has) trajectoriescontinuousHf increases MY ""L IBachelier ) outSt t1900 tµi = wnwmrh. drift !Brownian motion withModigliani }{St WtSo Ntexp rt: = er( )Geometric MotionBrownian 1966GBM binomialNitto¥IfhtemitGeometric limit theBrownian walkofrandommotion iConsideraas I]taken fatthe thisparameters Erstd Min I- pusu =way e..: . at- ii .,following treethe binomialConsider : t noto istst -?LsdSu✓I su \introduce f: SudjSCO) s=( )if Pu{ !?{ I a! . Sd'IiE cm- ...Consider Pu =P : EIFI (El EIIs" Elli] f) ] ]" dm Yithem Pls- ) YiYi mph )Van=P MP PVAN p)ft )go =-= =s .u=. a. )I IftIf !forTaking legthe takelimit thethat logarithm themimpliesst so wesom- - . ,,definedbe will completelynormal be byfom itswill andand variancea meanso v.vm :- . )(of all distributions limitifnormalSum thenormal inso →m -)( Eesti3¥ IE )( IfiIIby FIlog trust(d) crustlogos trustYi rust Vst Yit Yim= i- no -= =-, )) trustvisitSIIleg I tlimit Ithe slingfinsTake =p't -# .: .., .. ItV))5¥(( E )Istinylog FI )him I Vantilhast 'tplsVAN ta 'tPhhasthoist niran rst;- no-- =-- -- ,At o ,→ StGBM so-nsits)( )IntFIlog 't { }o trueslot itNn exp., .hove forWe like dwant ofsect thistohowdXt t dwtmeaninga an rtµ-: -?due differentiableWhat !It notisis IofRiemann definegladHGH thisfinite tointegral beshould integralSteffen total variation in ordera:- tforma !integral thatConsider duethe obviousdwt finitein !wt total variationnotis: aifo !alwaysitin it downfoes andup ?So Itithedefine integralcan you filtrationIIm )have (stochasticto introduceprocess { Fisttoconsiderorder F XtIPyou aa i , ,, ,)I measurable It that filtrationwant theWe adaptedFtthat informationXt thethe and imis in- means.stochastic t theuntilprocess is same{ stochastic861 process 'll )Def oillf E, El ) ]go845 thetor ,, 'dense Linare,Alf )Ctu constant tmtotutsevery isfFunction om Tsimple: o =i . -- ,, !constantconstant variablerandomin justnotf aBn a. . ,functions dense thatL2Remark simple approximateit preciseim meansare asI can on;i functionfunctionwant simplewithi aanI ! HIMHillEgf )HitsIto ffltldgltlintegralDef f -= =: I value beginningthethe attakeI,functionfltthere constantHehintegrated w.r.to winein processan.rs .-. stochastic:/?)the CConsider Xt{ FtlEPprocess r: . to, ,Def if ft Fe measurable Ft adaptedXtXt in is: -- forfiltration ×Def whichminimal Fecallif ItheE adaptedis weis: ,OJXt"" NATURAL FILTRATION )(filtration informationofflow "The "like the the Flowis information ?"}Xu FtfoggytheConsider { theprocess Yes} becauseeventand Afe. Xt A9 Er--. . .+.definedt completelytime the Aat event in . ?} 'event becausetheConsider Ft definedNo completelyfoggy A A notinA c-gr= u . .,.function .tnConsider ti bsimple togrid afa oven a -.. .... .,. ,,f Ihave ]by definition 7¥ Wtsi)WC )Isil sitswe ofdwyes , -,tmbitsto ,-as -Remark dense Llapstep constant harnesseswine inare: - , ¥} g:{!MedI T LiamIfRemark stepthe constant harnessesconsider fesswine f e-: . .here step constant Wienerhwanglad the processwine integratedis witN aB -. . ITSPROCEDURE Ca?L b)THEDEFINE ProcessFOR ATO INTEGRAL g E ,§ )( '} ds{ fldsimple1) fmlsof )EinmateA processes sitwiththnx of →gm -sequencea .④Substantially { fmcsigsit glass→. b! dm2) well defined"the SmhFor integral stochastic variableeach itis isana andm ,Esto there variablestochasticpossible that exists tprove EnZ zse as →asm.. !3)2)Propos i t onIs§ firms fmlsldwlslglad weststochastic3) define theWe integral --onnow ..Properties : b§ lab ) I) (gladness )EfEf ELSYSD dsglad theof integralWess ) Ito1) o isometry- =a- :2""! Fbwglad Wcs measurable)3) is -thei mProof whereof simplehnohhty gcaseI is : ))fb El lIIIEL )III) fftullwftkte wltkteEf SHH tu) Fewdefends)tie - adaptedgiddiness Since- isg=. .. ,]lo defBybehaviourvalue intervalonly the Wiener tiethethe ofthefftnl processon on , . .)(has fftrincrements Thus haveWftuIindependent independent )ofWHVV in weso :-* . .) ) )El Eff flutters)fltullwltnte )Wftu)) MtnHid )#Dawg.- o- - =- ↳ oto where Lto !stochasticfind Eat ]west f tse processinEBtn , a,. . ,MARTINGALES Eko EfxIf EH })Def if FsuXtD Eat ] martingaleifthedrafted I} ifis isr is eco -: -, ,Fito EfxIf EfxtlFile)ifXt Eat ] ifthedrafted martingale} Ifis Xsr is superaco File, ,IT EfxIf EH] martingaleifXt Eat ] if subtheodafted I} ifis Xsr is aco, , ! Few:#L theFor martingaleprecess It fksdwls)process isfany anE : -=: ,)Proof them(srtconsider :: , !!! ) ! ])ELEl ! ) Ef(duntSH I I FsuFitEw Fsu IWH fluidfluid ShawnwaswentsaidE t= = Fsu adapted)! -If !§ WHfluid Dtadwlulgladwas gunt ==hubertyfor↳ Io § !thatNow MaidsXt rdwlsl)No it tnmowwe =: xa IFTDM Riemann toTERMf){ d dwt!Ndt to differentialstochastic equation,G formsthe