Scribd
Upload
35 views69 pages

Probability Notes

The document discusses concepts related to mutually exclusive events, probability distributions, and relations in set theory. It covers topics such as random experiments, sample spaces, and the properties of fields in probability theory. Additionally, it explores the closure properties of classes of subsets and the generation of minimal fields from given classes.

Uploaded by

akashkamble9475
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
35 views69 pages

Probability Notes

The document discusses concepts related to mutually exclusive events, probability distributions, and relations in set theory. It covers topics such as random experiments, sample spaces, and the properties of fields in probability theory. Additionally, it explores the closure properties of classes of subsets and the generation of minimal fields from given classes.

Uploaded by

akashkamble9475
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
Mutually . ° Exclusive. Mutually Exclusive And Exhaushve, CollecHv ely Exhaushve, Ar UA2— UAto =-2 mufaally = AU = > S| re Penne oS l63}2029, + Mixture R-V3- X~ NCou) Y= degenerate vr. u with pov=J =) «oegenerale means | 2=xtN_— is mixture % meensel SESS © x Ne) Z=xtY vn Nin " axtbe N(autb 1ao") HW Take generale xaN(oi!) +500 2 KN) e003 Yn BernoulliCP) with some oxtb ~ NCb a’) sucess prob z2=xty plok density curve > plot(density (zy) ; i Q see it is mixture or nok? cor of diocre te. F(0)= AFx(o) +U-a) Fylay A €Cou) conven combina tion &> GcoeFRcient bet” cou), @ radiion of coefficient ts 1. b 1% for different Fea)=AFa@y tCi-a) Fy la) € Cou) nar b TF acb El ‘ F ib EIR cof of confinuaus " Fecay ¢ FaCb) Fuca) < Fico) (1-4) FxCa) £ (1-4) Fy Cb) > Fae Ftd) Mixture DIS 4 UEVvueeVGY DD DD DN NN MUNI EE DDS ling 2 iy, =| Nim Fata} =O UN Fata) = 4 aie a4 -0 hee Fea) is not af ao F(a)=O be Fy(o) =\ oF x Kk 2 ai=| K Fea) = ZaFxica) Ai E Con) oa L Rw write proof for cdf o« not? — ( Function satisfying all the condition : ists amu x 9 FC) is cOF of A! ( let FO) IR Cow) be a oF being Cor. Then there ex . Mixture TY Is ( convex combination oF any D+d, c&d ,cdc,0&D OV ' @ is i+ COF ! 3 aw take 2 aifferenk RV Cd0, DEL, OED Fea) = 4 Fxco) +(x) Fy Ca) ECO) plot in Rp with different dist? - a Relation A crelation From A te 8 jis a subset of AXB A relation from 8 te A is a subset of BxA eg AzCarbeC} s B= C23 Axe = Car) (0.29, Cbs) Cb) 1 Cet), Corry axe = { Cha), Cb); Ue), (2a) , C21DIs cud} JAxBl= JAIKIBI # Types oF relation @ReFlexive Relation @ Ireflexive crelation @symmehic @ Asymmetric. Relahon © Transrhue Relation. Equivalance ; " Ae Relation ~ Reflerive + symmetic + Transitive Partition of set? leb A=O Let Pe, Por Pa)---. Pn be a subset of A st @ PiOR= > wv it) @ bei = A Plo) =) eq OA Cu23} TF IAl=1 — pa =l T Cry, C23} Inl=2— Pear iT L133, C24 IAl=3 — Piles id £8 £24, £3} ! id Lu23% y Walaa — PCat) = Z RorPer) “Rand Expriment = r ao cp ise outcome at ang tal is not Known with certainty is not known with certainity anless the exp: Ss actually conducted. +sample spaces set oF all possible outcomes ~element of (2) are nothing but outcome which is denoted by wo. + Deferminishc Expé A crandom exp whose a is singleton set. *k Evenk & > collechon oF outcomes oF me — event is subset of 2 (Ais also event) — every subset would be an event? CNet) — In some me it is not desirable to consider every subset of 2 is an event. « Elementary Events Compound Events ~ an event which is singleton - CE is not Elementary subset of 2 . Evenb. Hie VwWeD - subset of .2 which {wo} is Ele event consist oF ot least also foro 2 +wo elements - occurance oF an Buenb ¢ If weA 3 event A has occured TF wtA 3 event A has not ocurred v ¢ weA® . ACL Ac has occured} = Note & at ang trial either A or AS occurs bat net both Beacuse AR AS are matudlly exclusive event te anat=o AUR RD + class of sabsels of -2% (onige:>) @ Partition of eg C= (AAG, ACR ees fA0B, ANB, ANB, ATA} ... A&B are not disjoint OO OO PO > Every seb A is a anion oF singleton sets oF its element > A=C23y B=fQu (av L3t 2 9 ts) > TF A is finite —> Union is Finite * fis count ini — onion is countable A is Oncount — Oncount onion %* o- Feld ocF*2- Suppose that -2 isa sample space A non-empty class - @ of subset of -2 is said to be a o fell on 2 , if OER - @ IF AEA then ACA ® MEA VY OFNAB--- co then U Ané @ nel —. A o-Rield isa class of events of our interest rrelaked te rrandam exp. + closure property oF a class $— suppose. that € is a no n-empty class of subset oF ~-2 lck © be any set operation on elements of -< she class ¢ is said 4o be closed arto the ope O» iF the wesalt of the ope®© is in € *ATheorem + (c- Field} Suppose that + is q «-Reld on -1, then Oneg¢ ® F¢ is ee closed wrto Finite unions. ® ¢ is closed w-rto intersection, differences & symmetric differences @ ¢ is closed [Link] finite intersection ¢ count inters” + TMeasarable SsPace ® suppose that -2 is a sample space € ¢ js ac-Feld ona Then pair (-2:$) is known as measurable space 4 A measurable, a numeric value can be assigne| to every set in $ in consistent way. u she numeric value that is assigned to ned will be called as prob oF A- te PCA) + Examples of & Field s @ie} 2 bea somple space Lek Zeca) , That is Fis class of all subsets of 2, amen & GF is afield Ce- Field) on 2. > Oe crpsa ky ALN, \ © suppose that A CF er FF is closed w-r-to corm plements @let (An] bea seq oF sets in F* ApS NEN 0 3 VU Anco nel *, , Fr is closed write count union # =170-2) is a & Feld on-r @sup pose that 2. is a sample space & r= Cor} then tis a a field on 0 > © de4* ® FF is closed werto complement beacuse gf=- & =H @ An=b Ww nala--- 3 Ao= phe f ts {Ang be a seq oF sets in f* In this eg there is hoo kinds oF seq @® an= > Wineahe--- Gan=p eH @® pt least one An is BS iY An=m € + Hence $F is closed [Link] count anion Remark & x Let 4 be ang «Feld on-n then, fect € Te J L ta} w(-) pore targest o-Feld on Ene @® In above discussion ,can we think oF some cluss which + not a o- Reld © c={4) is oot a Feld is not closed wrt complement @ C= {Oth AD} for Axed A S.0 is not o Feld © T= AB, Ar} not o o-Feld TF c isa e-Reld it must closed orto Anite union Here AET 1 GEC but AUB is note @c= (0416, 00, As o.3 ®-2is a sample space. Let $=CO AAS A} where A is fixed let $= {A.A} then $ is a o Feld on-a subset of 2, then F isa o-Feld on © det ® tis closed w-rto complements ® Leb Lan. N24 ts a seq oF sets in + cage@®* Suppose thet {An} is a const seq say A® -Cwhere A*= oO, Ad Af on 2) Sez Oats pt no PR = eG = fA et cage(® : suppose that { An} is nat aconst seq but the no of distinct sets in the seq are Finite only 6 An is infact a finite anion n=l © suppose = £4,213} Hoo mang giffarent o- Fields on 0 you can lisk?. i ™ Falher, too) SSCS Fo= to, 04 a3}, % tue (4, toh, tng), 23 ts = C4. 033, C1}, nF ®@bnion oF two o-fields need nob to be te TF $f @ f are two = Relds on.) the fi U dr ore qeoec-Relds on 2,4hen need not bea 7- field eg tsufu= £4 OF, C23, Lady, £1633, 24 is nob « Feld Tk is not closed wr to union a o- Feld However @otersection of two- o- field Cany no of o-field) is a o Feld te iF th @G2 ave two > Fields on 2, then t= Hoh are too & Fields on-2, then > F£Osdnce fiat aw ~ Felds on2 = 3 ¢eh ¢ eh E 3 o¢feohe= Ft 5 @ibhe to Heh 6 eh cea de Ace = 3 Aten oor F is closed > 3 aehoh= t w. fo Onion = @ tet Lan nt1y be a seq of set in F 4 is also A ie Ane vnz) o eld en DSAnch 4 Ané tr we nZ) = > Bren 4 aed. 2 Bacet le we ee @® suppose that -1= LOnwy---3 ds q countable infinite sarapl space .let $ be class oF subset oF 9 defined by : t= fren! either Ais Finite on ASis nite} hen the Class # is not a & Feld on Remark Both A& AS cannot be Anite . iF A & AS both are, Finite 3 AVAS is count finite bat we have AUAS=.2 is count infinite 7 Osince bisa Anite seb bes @suppose Heft 3 either Ais Finite or A is Anite case CT): IF A is Finite \ then (ADS is also Finite ' a (ay et > feet aseCh): IF AS is Anite a awet , 4 is closed werto complements @ Here -2= (wi wr-- a4, Oefine .An= (Wend n=li2-— Ar = (aj Ara (wy | fe All An are singular sets & All singular set is finite seb ws each sn is) Bite set ¥ Oe1— But Uh= {Wa,wa--4 is not Finite seb nat : and ike eomplement , EN ic An) =(edws,--.] is nob Finite set. (o 4H is not closed wir to countable. unions. Hence class 4 js not a o Feld over m- @ suppose. that 2 is an ancounk Con) / Core) /¢-09.09)) . sample. space..let ¢ bea class of subsets of A st t= {aen|]either Ais count or AS is coant} . OF “his Finite set @ Let Aet » either Ais Starz js—counk count on AS is count caselD) ! TF A is count ave oe > (at) is count — > Aces case(D): TF AS is count | A is uncunt a ace 4 is closed w.r-to complements @let {panz} be a seq® of sets int ive Ane Ft Yo casecZ): TF all An are countable 3 An is also countable, - n=l case): At least one An is dncountable 1 Say Am since Am € + 3 as must be countable a c 5 A Ane Ar 5 fad is countable .. (subset of countable seb is = countable) 3 (Paty= Ban ef Hf na) at Hence F is a Feld ee | ‘an *MMlinimad sigma Held + ~ lek © be the non-emply class of subsets of a then the smallest o- Field cantaining Claas is called «Field generated by class. or minimal «-Field overclals c [NotationeCc) | o(-<) © TF 25 C-0.0) then a= Luash ef A= set of alld efF A= set of all IG et AS is count B=) ¢F BS= C-090} UL100) is uncount + The importance of example ©: Here $+ 1P(-2) beacuse both A AS ate uncountable, * Points * Jw countable infinite ya- cont inf 3| multiples of NEZ > cont inf 4) @:counk inF CRational) Sjsubset oF count inf > is count infinite g| set of all IR uncountable, J set of all intervals of IR> oncountable, €] @° = oncountable 3] if A fs cont. inf = AS is uncount subset of IR, converse is not true) dk Minimal o Feld [ smallest e-Reld | Generated Feld — suppose thot -2 is sample space let € be a class of subsets of 2. Now. € may nar be. o-Reld, gut best of our knowledge ‘& consist of known events of our interesb class & = (AJ There ave SO MANY There ave so many =-Feld = Fields that contain that do nok contain & vw eg (Ory eg Co: A Ad £018,0.% 1 BFA \PL2) L we are interested in smallest o-Field that contain a given class © eg if ‘w= Cay dhen o(©) is the smallest o- field is xe) = Lo..2s A EY sDeF* suppose that € is q non-empty class of subsets oF 2. cwhich need not be o-Feld). A &-Feld $* ono is said to be mitimal o- eld containing Bi iF msn: Cod Msi: For any other o-Felel 4 -that containc © , $c G is denoted by ote) * Properties oF o(-€)* f i al @ cl) is a o- Feld on -2 4 @ ecole) A @ cet . where $is a «Feld a 3 sceycek 4 By assigning prob to sets ing class ©, the prob oF a sets in oC) can be determined , if -C is appTepn rely ¢ chosen © Lemma’ suppose that Ci & Te ave hwo Classes oF subsebs © of 2. TF Ci @ Colter) then Ce) ceCez) a €9@ suppose that AC tet Ciatay € Ce= tay then o(ti) = oC C2) > stepCi): TPT eCaye «Cer) we have Ce CrCcy) 3 At e(ty prectieye oe ns Retd] Hence. Ti C oChe) since 1 (C2) is minimal sigma Field ‘that containg e, we get o(e)c «Ctr) na nme mnnannanannnge step ci: ppT eer) cC ete) Peed we have A cote) ac cola) 3 & Colt) 2S sth) cre) Hence. From Ci) @Uii) ete arlcr) FP @suppose that A&B are 4wo sabsets of 2 eb w= CA 18} IS gy -e2=( ANG, ANB, ANG, ANB) me Then (1) = ©CTr) 9 Theorem t . p> suppose stample space -2 is count Cie Fin ov couk inf) FD lek cs be the class of all singleton subsets of 2 ‘oa Fe 5 = C twh] WE 23 ‘ee Then &Ces)= 1P(-2) 7S o> * The Borel o- Feld on ik 7S suppose that ,-€a is the class of all intervals of the 7D Form Coad, wer => he Cy =Chom a] | weir} = Then Cee) is the minimal = Feld generakd by class = te is said to be the Borel & Feld on ip — =<, — we denote *(ts) by BUR) | Bie |B Sc, - The sets in the € Borel o-Feld are known as Rc, Borel sets 2 Sa Theorem t suppose that B is the Borel = Feld an Im => > : then : ® Bll unbounded interval of IR are in B ye For ac ein Ce. %] 6X60), 00.) LX 100) are all in 8 @® all odd intervals of IR are in B Fe Qeb EIR adcb Cavb) , Corby, Carb), Cab] arte all in B © A singleton subsets of IR are inB te for aerR fal eg CbbbEbbdd @® on Atte on cont inF subseb i are in © The complements of alt Finite & countable infinite seubser! of IR are in B, : + Theorem & suppose that x: LI then, ® Cre Qe] = ba @® Crew] =2 | [xe UBn] = ULXEBN] .. whe Bo ove subset of IR Atnverse image of a anion is the onion of inverse ' images} ® Cxes]= Cea’ { inverse. image oF a complement is the complement oF inverse image} | ® Cxe OBq) = OLAEB] inverse. image of all intersect® is the intense? ' of inverse images} ' e lemma é oppose, that XLII, then @ CxraJ = Lx¢gal® @ cacxeb]= [xe b]- Cx éa] @ (Cxeb] = Cxtb]- Exsa] @ BH Lak congatk) = Cxsa] © Cees CCxCe.) IF Bic By # Inverse image set ¢- aisleor Le} X! Qirm2 bea Function teb SO22. then inverse im age of s under X is defined as follows Keays (wer |x) eS} e igebra oF Tnyerse Image? let Xi 2D be a Funchon . Let Side. SP Take WEX x (s‘) 4 3 xlo) €S© " - EI C 4 2 xo) fs 3 nwes 9 wg Acs) > xiwelxe3 ? dee crxkewe ot @ we [Xl 3 “wEXe) 5 xe) ¢9 xo) es? 9we XCD 3 X (st) = C89 > le Ppsuse] = Xerve ca) 4 Take We K (S082) 3S xlw) € 50S 3 x(@) ES) OF Kw) ESe or Kew) 6 AOS2 uous 53 we Red or WER) or We X'ase) Ss. 3 we (5) ORCS) “| ot Now wexGvd or wex(Sr) ar WE 5052 C Both) KU@JESt ae Xl) ES2 or KW)E SOSt KC) ES, OSE 3 we X's USe) BouHuU ad? vt D ¥ Cs) at Cx'csevJ -) - at © *% (y5k) = JY CX'Ge)] 5 take we XC YS] = xc) -E oS > xw)eSr frat kast one K , say ke > KC) E Ske 3 we Kelsie) 4 5S wv one x (Sie) - oS BX Ge) ao © ox} £ cu 3 cov) 0 = BA = of least one ki Say ky a 3 nw) e Sia 3 Ko) & US 3 we X [Se] x ty yw? xts1) 9 x es) Take we k (510%) *(W) ES NSe “ewes & K(@IE Se 5 were) 4 wer (Se) 5 wexcs) @ wWEXG) 3 were) oxy) now ore Xa) 1 XC5) SS 2 we XG) & WE XG) 4 > xw)es & xCw) €Se Pap Kw) E182 » 6? we x Csi9sy Wo@x Lise] = 8, Xp] _. Fomn@® = since Sr Sar--- Sn © 5 sh isf .-- sr using @ ar ostl- Sree] eLtgedd= i Lea [% cose] pacxtrl: 7 eel FP x Case = 10 x [se] wow : wu aw ow wu ea =f 41s] Kes} = Lwem | xtor es} = {ae} xLrical) = Cx wer'eor) 3 xe 5 xxisy] cS always a: xLXes)J=S IF fa is onto ® x'cxcal ¢ A only one, @ TF Si &e ave digjoint then rag & < lel side ave also disjotat 5 eb us suppose hat Xs] & X Cex are not digfoint 3 wet BF 4 we Cx Gd) 0 (Resa) gy def” oF iaverrse image seb 3 xe) E51 & Kw) €Sr 3 Kl) € S105 a Si0Sz> b this is a contradichon ro) 5 or Sze iol) w~ ©uwuovwuvsv ®& Beacuse Sasi Het common point gas wir pre-image. at N x'er) & X¢52) Hear tt wet ae RGSNTAAT BEE few precing oi Wer sce) = Ce} siz CAG} xiey = fared COsy sence ik is not necessary if MTs & ¥'GseJ are disjoint aren A si & Se are diajoin IF gCs_ then xcs) € XG) TF sicSz, then Keane KE) Take. we ¥¢51) > xed) € HCSr > x(w) €S2 we Xz) Xi) © KG) Z we = Now, IF ¥C1) oar soo A siz £2133 dea | “3 Ya) = ec} Sat Cas} x =" | X'@)= Larbiccdeey Beacuse 29 Het Hence converse 15 not true, aaa point @ den Fea Far Ainvase image of a collecton oF set Lek = collection oF some subsets of 2 Roel = Crenrfetateg- = (RG) /ser} averse inoge of a cfollechon oF Oo a sek is collechon of inverse Image O eg “Tossing two coins coe CAHOATCTH TT xe) = No oF beads K (4a) sa XCAT) = XCTH)=] XLTT)= 0 cor = Loiiay = (004, Cred, toy) xtc] = t x(Cot], RCT, Kee ot] =¢ Crt} Cha wT THY. et nna} Resalt © iF —€1 Cea Keds Fey lek 86 XE ggec 2 As *'ca) 3 since TU Ctr Sé7Cr also >» 6) ¢ REes) Ac € Leg mage of a «= Feld isa e-Field XPV q o-Field of subsets of Qe is ae Field + Inverse | ~— let IF be sp prove: x LIFI take Ae x LIF) 3 4eclr oa XLS] Bat IF is a & Field sein 9 ¥Ls] € XD ace -KOF] xcs) = Cece 5 9 oF 21 such thet Let EAmn zi} be a collection of subset An 6 x OF) v 92! To prove aS) An & FLIED \ jon cir 98 %ES0J = An usa GIF 3 X[uside x DE] 5 Uxlsa) € xo > vAné LIE] 130g] -2022 _ ES cr Ste % XCe) ¢ XCce) © Te be two collection oF yg, be a Function - Leb t let XK 2-2 subsets of De > Cc Ce | Then = xeay= kes) see} X(er) = Cx) 15EC} vek AERC) SGT BF A=XC8) since TEC? this Sércr as Well 3st 2a A=Xls) sample S a ner (ee) Aven cxtcr) | (ee vas \ eS ees SQ ap xCe0 © KCC) x acres ase THMTOATITTS . No oF heads Xu) = nat fonusy r= 0 Coy, ries tot sabsef+ ce = Teh Oh, CH] geen Kt fT}, CTY ices) = CTO CHB HHS PS lek Ki. 3 2 be a Function LIE IF is q ©-Reld on Subset of 92, then CIF) is a o-Reld on subsets of -25 9. IF A isa e«-Field on sal Ir = (SC / xcs) € A} s then IF ts a «Feld & XCF) CA aT bsets of mu @ iF BAF Cet 3 Xa) XC) PLE LIL EE et Thm Let Xt DIA me bea Function Let C be a collection oF glbsets of 22 Farther let o¢c) be the minimal o- Field generated byt en > oex\= XLecey] so Let Re CREST xi cCetey] — since oc) is ao Field , by the eaclier Tesalt = x'Leccy] is a «Feld = e+ £6 CC) Tid XE] ¢ XLS Cd] From above mesull © = using @ R[ecey] is a o- Feld containing XC] — since 6 LA'Cor] is He emallest o- Fele contataing \, AT], we get — oLe'teIS ATs ce¥] ? wo prove XLece)J co XC] = eCX0cJ) isa o-Feld on subsets of -Q) A Let AF ; £5C-02] x05) €o Cac} pays .., KF) Co CGI) sy can we prove, oC) SIF Tt is enough to prove that “Cc CIF & b Take SEC 3 a! xs) € Ee] ¢ o C¥eeI) XE @ CK CI) = SelF SIF o Cc) SIF 3 Xlecey] s XCar LIF is o- Feld from cresult © z0]o4)2022 a= f-2,,AA6,.09, Bun’, BA} a sResulk ¢ o (aces) = Rec) | IF is a o-Fiele] on subsets oF Qa /x CIF) isa o-Feld on subsets oF MH Ire (Suse--3 XDFJ= (x c5i), XC} e-gO N= CHa THT TTT No of Heads xt) = i= (on 3} Ze) = {TT AT TH} se (343 Ke) = £6 @® 2 CH AT TH tr xt) = No of heads @=C Can, Cut THITTI: b1-2F Ceecie) ofS, 1451 aes} fsem|oesles.2¢s} ag|oz)20272_ 7D Random variable & —S Leb X8 CA AJACMR UB) be a Function ~ . TS Kissa mv web CAFE xXcsjJ ea v[se18 > . TD Economical oefaibion:- TS Leb xi car Al (RIB) be a Fanckon « lek € be a class of é subsets of IR Df TCC) GiB, then WE Say that * is av ~3 Sx Equivalence é 3 | Suppose that %8C-2 A) AIR 1G) is Q ow then we say that XLeceyJ= e LX] wets CUA) TF ATs] Ea ¥ Rw wer ere ae inverse, ack ATED aay \e-at subset, ag 08 Per oer” (& Random variable DeF”). 3 To prove t « isa ew as per veF* @ CE@momical pefinitien) ~3 3 start with ony sec we know that Ce) =1B € CIB 3 sél3 XEsJ ea since ¥ IS A MU as por oef’ O we have proved that x is a wv as per der’ @ gs well Oef* 2 = peFIL “8UREELUE suppose that Xs 6-2.) 7 URS'B) js q muy weeeto Der O ve gece W set, Xcsyea ot oCe) =18 To proves XTs] €A vsew Sug) ecA Givens NLe}cA A isa = Field SCR eq] CA by def’ of minimal & Relel S[ece] CA -- by X Lecerj=e CX’ LeI] Exomple = © coe fan HWTHTTY et tHe), CTH, CaTITHD, G2) CANT (Th HT THY, PHT TH HH)Y y= No of tails x20. if W=HH yf © = AT oe TH if @=sTT o ofS 14s, 265 ca} o¢s ,\¢s,0d5 Cut TH], ofS 165, 26S {17}, 04s, \€5, 265 THHHT THE 0€5 1 es, 2¢6 CTHMT TTY O€51 (€ES, 165 CHAT oes, 15) 2€5 aad OGS, 16S: 2é$ ata = bob b bb bbb ddd dddiduddddedd Pe BP bObb EE c= Coa) } aciry zi oo XC (-10)) => a 4X0 tp = fo} ofc oan Sh wz) xJ Q@isxa Funchon. of \Ki? X me x Is \x\ a Fanchon x7 Vy a sk Borel Funchon & xe aaa a 6 @ Leb gs IRGIR be a fanchon we say that gC) is a Borel Fanchon iff WSelG 3'csJe18 gca)= ~™ g Crd = > iFoaco OTT = [wa wal }f azo we) of a a ee eacanale g [rma] ea a gc) is a Gorel Function cay economical peF"] ae Economical ek of Borel Fanchon ¢ Le giypaiR be a Function. further lel -e be q collechon of subsets of IR OB &CC)=16 then we say 4nat gC) is 0 bovel fonchon ifF vs ee, gtsJew ProoFt- g’csJ ela VSEc ate ce ry ot rs 3 a a] c «C§'Ccd} ce 3 § Cr Leg) £16 oejc ie This is part Tt tai me ew eee ee ee eee ace eee se Borel Fanchon of 4 vandom variable is a ov Le} XE CAA) CRU) be A wv atmell Sarg SE 8 Farther. Leb gai be a Borel Func bioy nt e co co” sen aexy s CORP) is a ON To proves ' Cgcaiytsl € A u ° FLAC) % phe Cgeytat = wer |g (xo) Es} mposite Finction X¥ CMAJ ICRU) xcs] ce " {wen |xcwest (gen se x0 : = {wee ] Keo) € grat since gt-) is a borel Functon §'ts] ¢18 =foes] werlg csi] } = xcgcsi] Hence XCGesy] eA Hence g(x) is 4 random variable . » 3 Hene Bael Fanchon of a mV is a ov rksigma Red induced by aN] Induced c- Feld > ‘ Lek x2 COA) RNG) be a TU Farther , Leb XU) be defined 05 ¥Lil= Lkts] | sea} snen XC6) is known as +e-Reld induced ty x ¢ wk is. denoted by ox) RV : : oCr) a3 mtirrimerl Zady | oC set) 3 ande minimal 30] 08]2020, Results Suppose Ki(-2/A) CRIB) is amv Farther quire is a Bere Function Then, (glx) © Cx) Proof & ~ teh RE F(9CA)) 4 3 ne[geal Cie] : a etx): AG =f ACSI | scr} 3 Jsele 3% A= Cgen)'(5) =xL9' cel) S is Borel then Gy Borel Fu? gt) isa Borel Function -~~- — : KLE] is. also Borel gis) et az RCS] where Sieg - - AE ox) l a Example oO x2 CAA)ARIB) is A TN gta) = a Sate -c= {ewo)]a ered a=-\ § Cem) = {xeiR] er) oy __— = a A Fema] = 6 — aeons” ZLewell = Cas] E Glew. dl = Caubr} anole For Some anby ENR a 3 LCeal) = fs} g=-00s ol By economica) deFinition 3 [Link]] E18 Haste ge) is a Borel Funchon . ack root) nw For quabic. equation (try t0 dish ® gex= sine =) 9 Cem a] = 4 qe-y a ays = Yaar cana yr q=-1 cane Canty we bbe bb G&S LUO UDO DD OO Ieee] = ( CO Jno ° | U {ont |n=0 va] a=-] io Mi =~ B Lewd] “1a Borel Sehs on ay) C oa ingl yV = DU Lai . © single ten gcennd ie s mes AR ae PRre @Ocl mw et © countable Hw Evegy continuous RincHon io a borel Rinchon = Cry to Proof] xReoults, © IF C1S€2 then TCo) € ole) + 1 Sea Celta) 1 & elt) e(c) ¢ oa.) — oY definition of minimal o-Feld 4 Random Necton ¢ | -ev= {Caribr) XCe2be) Javbiaubeery gy yy . z af car [| o 20, 2b) cand) x Caribe) ~ Ca. Zar Lj PCx=s) G2) x8) : i sn ET interval 3 (s? xe = en | xodeciF [necna}= {we : ¥y aris saer = =—02 RIT commen part GaaT aaditien AA 3 ECMz), Ke EUR > LH 2) % etre] 2 Cxe a= Cx=3)02 = CmeG) x Ele] aca herd a at & > ‘ 3 (xa34= [x3 y [x=34=3) 0st DS Lee sigs Die Sins Xs €tRe, ma Cl , -- Wie EUR , Kit @Ru-- Xe C1] az xi) 2x (si) R= (KK Xie) to . Kes PwEH/xicares'} = (wer/xiw) esipoRal--. az ee Ck-) mes = fwen. (KiCoIESi} 9 C09 6-2 /ni (0) EIR FOL Mwe-2/ Xi (wele} @ on $e wER /pi'OIEG9.0 [HWEA/ Kec) ER = {WEL Tico) €Si, Kice) EIR, Kelead) E-- Kirt KO7EIA, Kin (OI EIR. Xecew) ete} = {wea / x eo EIR, Xr EIR ye. RAEI, RLESi 4 Kit EIR, Kid2GtRy -- = xu eIr] = [wer /Kio) EIRKIR ~~ IRKSINR-~ 1%] 2k LiRKIRX-- XSIAIR-~~ AIR] Eo CX) CX) CoCK) ULecxil] ocx) cL oCstxiy] colx) wow cose O qe prove o(UscKi))> ocx) Take Aeocd) 4 skeet 25 a pis bation Fan clio & AF) = seb of points of eliscantinuiky for cor F scr) 2 seb of points of increase Tor a cor Fl) pernoull = cee toc i y TCR) = {or 4 Coisconh news) Uton) ¢ dEI= > 1cF) « € Low) cof oF UloW) at ig a point of increase WE Feath) > FOX) 79 Ss Be or Flex) > FCx-h) Ceontinuous) ae now support oF % piscrte mU OCF)= 1CF) 2 support oF x (ontinuous MN OCF)= b , TCs support Hews Real WiFe eg oF mixture OU peo hi lat) ha = Not) n= loo © st take poy From Bornoull) — Kia) @® —=r—_ from NCou1) Cui = Stoed 0% Chances comes frorn of fot ‘ Co%e ay C-4) Fa. © Tne Z= Bernoulli ¢o.@) Final cud Cw) = BernoulliCo-4) w <2 THAT Bernoulli BRIGAT x BRor d=o4 oe at Na Bernoulli Cv) Faltel 40 %e zero xa anda gay Gotle one Baler Xe guia EME @ ce sponding ost RX xis OW NS Cwie 2ixitcezi)yi J x z lz lot XY a te mixture eu Aa oY qa Scte_EE ONET Jot W { zw Bernod\ti X)| SRF ator Go 2 2 12. jog} 2022 a Li Dimensi: Udddddddcdeddd It DS pew construct iRKGIR or take seme 1 which io ™ SS ona) to one which it is not measurable. i %& COF GF random vector t a = == 3 ae Properties of COF oF R- vector Feuxe (Qud2) = P[xi¢ar,%2£a2] a O OS Friyxa CO d2) £1 3 ®t is right continuods wr to <3 both component 9 @® lim Fx (aaz)= 0 a. > we COF oF Random vec tur <3 lim Fx Cova2)= 0 + F Cayo} =0 3 oe Ror a2 2-00 SR oR <3@ a8 Fx (uae) = Fala) Ree ae 3 4 ads. > + Fcacoa) = Fra) > fwe 2 / Kilo) Le} er set Yt 0. -drerre A on xc) Loe = Facar) 9 IF cos is net increasing ? at least one pale Fg -arb €IR a FCb) 4 TO prove This is nob CoP 2 — J aib EIR a FCb) PLxeal > PUxSb] PLxéd] > PLxeal + PLacxcb] o> PLaexev] OF Cobo d= FC bi G2) - Flanba) + Flaua) 20 Carib?) Cb.ci9) viog OLOO € (bi be) Fea raaje § adder -ve Baer ¢ sect anraeur ¢ Tes @ Fed Add Hear aretat , 6 € & s « For any cumulathve dictibstion Funchon set oF discon # nuity points is atmost countable. —, Let > be set of discontinuity points eity—size oF Foe) Farther , Leb Die= Set oF all discontinuity points csith size of discomti nutty greater than = Da: set of those ~% of which the size of dis continuity is greater: than 1 a th Da size of discontinuity at x is st > @« is a point oF dis continuity =) «ed [Link] Vk Ue SO Now Take XED Prove by -contradiction suppose % ¢ UOK 3 CED Fro any k 3 size of dis continuity e+ Ye ein Sk of & 7 Re - 5 Sx is non- zero Se. < 1 us voKEIN S Ks at oy kein Sx — is combraclicton Carchimedean property at the set} OF natural numbers does not OPP: > KE Die for at least one k This ne eu er bound % E00. D= Ve Example oF — decomposition © FeCa)= 9 KEEL Oot eee 0 met L ance ql Sa e0 = ze we 4 - Ly we of me) ROOs EeO- rr) : 5 > 7a , 3 4 = \Saea? = Oo 2XEt = 0 1440 ~ 1 aren oexe eal | en set oF di's continues pt oF increase @=-1%7: t= Cxlogace i} {ene ? pi Xie Vo") XxX is @PLm=-]= PLXas 47 = pe Fatz)= 0 «<1 16m eJ 40 as N70 Wi wr wz YH Ws We O7 We Xo E- wi 4s ws, We Bound war we (® in 43] cg wer ws, Septea i fc in Rl SSer n, eercaeaer Set aT arr diff Bourd “wey Barf peb a GRE [@ in Fy] almost all points are -cormes “in ‘bound oF Carwor x) P ye) wR wen golto)= Ith iF @=H xe d =o iF wT = & iF wrt acta) iB wee a cay iF wrk = \ pointwise Conver nce & | Pintwise Convergence ) x (feck ave ern - __ |, ___—— | Cony ongence in probability | * Expectahon s- Expectation is 9vg oF population | population mean EQ) =H vexs EL {x- Ew] Eas Eft 2x] =h si-kF eee aon < cout) 2 ELOr EoD) (4-0) = od = woyxtH= Ee) euys ELK «ale = ELC ECO] + $c = € Ce) ‘par = efs™ se Cot) ae RY ne Cond} PLinn- xl <€J kato) = ©" xtw= 0 iF w € Cou) nie) : | (2) =) (FF wel 3) ae etl)! Let We O-1 A! geo | Ww YW= 60 yoee Su vegan ; ag yates 9 AST rr . nto) iF w eCou) comet | Xa) AYU) peintwise Fron Hate | xato @Sy xo 1. pointwise eeetay Result : Ow %9 3% mE then XY TS ; « Example 09 poinhsise canuengence & Convergence in Prob ¢ Katee) é wont tg tet pceg= ° ‘ PLCSJ= 005 BS é Xntwo)= 2” ‘ {xn Ceol) € wm ke XB (convergence. ‘ oA por tl we ae gio ar 4 i Seat 4 = et 4) (4 uN w= ey @) @) GP oe frroey O € oe oe rye apc es) wt gy arg -- 7° ie 1 ¢ 5 =o forse wet 8) OG xt 1 a) G) y= 2 ‘5 ) 4. 4) for every O wk cy) GO) CP-- 79 xncorrnc@) = e) 5) sJ “~~ KaaX pombe e ee ET construct Any @Y is q- wu for which we say Some person say For every um called O insted of 1¢ 0.01070 { Ss kate) oN@w) iF wekgrge xnlo) YC) iF wl ar wel er seq (ote.0-€) aie interval Wet Berd A seus | OT pointwise conuerg~ Groph Feo Poinhwise cnvergence_ convergence faze in Prob new gyre for n=l Lo| 1 xaceo) V1 Fre) @ 7eCr Example Tee {tks bt} Kol) = WO? Ki) = Wd Fita=o iF ack = oos if $44 = os 1 = 0-75 1 1 now Feta) Xelw) = 0™ ee os, Fata)=o if Os = 0as 1F as 29 Se s'as if Js eqet 6 76 =O" § = al 4 ee a 2 aos iF 5 £054 = iF art ts NoW Kolo) = © iF a ¢0-000000184 ame Fio(a)= 0 = 0.05 if 0-cve0lea4 ¢ g £0-.0009° 903 = 05 iF 0:000000903 @ mea sampk@ WW -— Weve Bm xd) - SAL t 2 (eae jake soo samples re N= (ove Ss wae BERS | we i % = Pa Bee Xo = — Ktooe 1 ( = — Seoo = Re = XS Koo Emporieal gwhot is empirical CPF pp nob, 191-23) BIB rar ois’ Hele nage 19 increasing olor ater ES we Prcang emporical ¢ OF vgs Oks VEE ES RS! , <4 evr cay = OFS n ecor ¢-28) = no of Ris “0 ¢ = et 6 oFC-os) = & v EcoF (-0-£) = ecor ON) = = 6 ccor (rs) = & 6 f ie Er (9) = = ves 1 c Ernperical COF is estimator oF se Convergence in Mean ¢ {xn zi} Assume that ELIAnlT] exists for every a % Sx EL \m&-x[*]>0 E L\xn-xl?] 40 Pointwise convergence Strong Limching weak om . is degered \ disk @ xo) = Wt Troy @) Xecw) = Wt Thou X%ew) = wt L pot) Aqiw) = WD ® PrKed= + P Lhe o] = int X=0 XD = Leweo ty VO PLxneil = PLCuol to) = = PC coat) sit n r Properhes oF Expectation . 03 [10 | 2622 3a non-deoreasi'ng eke eee {xm asi} oF non~ negakve simple anv converging +0 % ECKt) = lien 3 = NO Erin) ECKe}= AM E (Xan) | — By property of expectation oF none negative TU-S ECXin) 20 VN 3 lie asst ECKin} 20 ECK) 7 0 @ ECwA K2) = ECR) + BCR) as ECAN= lim E (xn) Kn 3x, noe xen Xo. ECRa) = lien ECan) rin than Xb Now Ain SKicnH) 77> non.-deoreasing seq Wo Fan < Koco) vn © Kien) tKecnh) ¥9 non- decreasing seq of converging te Xt ke _ min $Xon a frnrkea a2t) is a non-negative simple rN * Now ECKItXe)= Air, a)= MELK in} Kan] ; = 1M ECxin) + lien E Kan) ’ n-100 ne = ECX) + EOa) cE(K) Vcéle ® ECcm) cose I] czO . < ECX1) = lien ECKin) Ain s Kren) ™ Cxin € CXient) where fxm, az} isa seq oF non-negative , Cexinsazi} is a geq oF converge cx ECex1) = lim ECCAIn) = Clin ECAIn) = C ECAI) eo) cer) - ECC x)*] ¢- 6 Ce-xy] xtex .. X20 (-xy"=-% if -xXz0 et ae = 0 if -x40 2 texts -x iF xo ECex]= -€ Cex] = — Cc) ECx) case I) = CECK) HW Elaxr+bXe) = aGCx)+ bECKe) He Now Aré Xa are arbitrary MVS © ECK+tXe) ECx1) + ECxe) where (wen /x(o) BO , Xele)X0 Ooms KtKar%oy @ faz {we / Kul@) 70 1 Kalo) <0 nitKe % of x1 (@)ZO y Xa(ty) £0 XitKa Lo} ® As= {wee | @ = wer] Kilt) £0 /Xelw) 20 ytK2 20} |x Ce) £0, Xalw2o" © As= {wer Kit Xe eo © fee {wen [xcteo) 28, X2(WIZO xb xa 604 Doe ANA co Nes beacuse An-- Ae ate disjaint sets w Ae makes partition of or not), @ ¢ é é é é aibto ‘ ¢ ‘ (oe ‘ go. walat 16 expecahié ecx)= fxar § a é integration of x over entre 9 with proba bilié density Bernolllli 2 € é € é € € xdP = ¢€ (fxr x19 dt + SxTa coll a é safdP tar el? a af é € € = ap ca) ta PCE 6 6 6 lS E Cx tX2) = Ze CCX+Ka) Tai] 7 6 € ECxitKe) = Z E CCK +Xa) Lai] on Bt © X20, %220 3 KTaO , Xela 20 ECxiTm +X21a2J = EC xiTar yt E CxeTa) on Aa Kiz0, Xa Co CK Ke)ZO XilA22ZO, KeLAr Xing + - Xe Tne % = - CK bXe) Lae E(-xvIac) + E(-X2Tac) = EL Cx Hee) Tas ] ~ECutac) - E Cxo Tac) = - E(Cxitx) Tne] => EC xiLac) + ECko Tae) = ECcxitr) Lac] wence. ECxiTai) + & Cxelai)= & Lexi tke)Tai] wisne-& ¢ c 2 £ ecu tas) +2 ECxetai) = 2 EC cadxe) Lei] i it =I ECan + E Chel = ECKI+Xe) pr x is an arbitrary cei eee ecen) = E Leexyt} - E Ceex1] (ex? = cx if KEO = 6 iF xéo =clx*) (cx) = -C8 if x <0 = 0 yF X20 ECcex] = eCeceyt]- &Cecx)] = CECxt - cE CK] = clecet)- ECx)J = cECx) case} : cxo Biiedr = Cccx)t] - EL cexy] E(-*)=-E(r) let Cc=-a Where a70 > ELexJ= Elkady - €Leany] c-ax)* = -ax if -axzo 4 ax¢o a2 xd =o if -ax pcnus) {pca) + PCB play Pca} @pc =1-pcAl - PCB) + PCA) PCB) ripen) - PCO) Ci-pcn)] = ci-Pcn)) (1-P 8) = pcae)- Peo orn pLe-h8] pron canal _. pee)- PCA OB) = pce)- PCAIPCS) = 7¢9) Ci-PC4)) = pce) PLAS) Peaned= Pca Pca) © core=0 e Joint pAF= Precuct @ oF marginad Digjoint event ¢ 2 independent © wer ‘7 (0.0 |orcawe “Wr PCA)#O $ pea)to FEIT inept can't draw digrametica \ly. TOP TDHAMDNDND Ww routually _, Pairwise indept jnclepb& ® & Parrwise + mura ly indept indepte Prove oO: counte® eg e é ‘ate ala & Independence of k events * Independence. oF -K events ! suppose thet (An Aaw- AK} isa collection: oF events From (.nulFip). we say that this is @ collection oF indept events ae Kk P CORI) = oT pcai) > Todependence. oF countably many events - let LAnlaz} be a collection of events on CAIF.»?) we say that this is collecHon of indept events 1F P(A Ric) =F p (aie) vK @ Every evenb is independent oF 2-4 4 — Let A be any event ASD PEAS P(M=1 pCa): P(-2) = PCA) so p(an-2.)= PCA) Cand) = Pit)= o= 0 PLA) = 0l4)-P(A) ® Every even is indepenclen F of p-null event & also of a same event: @IF ao event js independent of itself , then ee probability is either zero or one. > tet A be independent oF itself PCAN) = PCA) PCA) Pra) = Cera <2 rcp)=o oe PCADMI Tp === @ TF A is indepr of Ae then pEA) is either Zee we one or PCAMAS)I= Pcay-pCAD & - PCO) = “POAS PCA): PCA) | O= PCA). PCA) & = PCA LI Pcar] Ms or ati “Inclepenclence of collection of events: e let €: dt be two aallechon of events + ; Eat =o = CalF-P). we say that the collection 4 © 3 8] Geel ew “ @ Ca are indepenclent of each Both sabseh —= cihen oF events wae & PlAina2d= P(AN-P(Ae) Mee indept ON Ae Ge Tr event are - # Independence of crandom variables Searle a Let x&Y be feo Tvs defined .on then bath indept_ COIR P) we say that x &yY are indept ~ FE @CK) & Oty) are independent oF Prob They is s Prob aS ameaswe each other. ve peasune. we Know teeny. ecx) = {Xrs] Jsewy = xCG] Kira = Pwer/ xt) ese} x&Y are independent Cgoint Prob is produch oF flargina\ prob) e(xesr, Nesej = Plkesi}. Pp Cyesy) s Ee DDPR MAOH AD PED a Fixe teste pLeaesil o pest] = PL (mea [ xem o (wer / Yo) es4] = PL¥ta) oY tsi] = P(XEsa) Ped) Xc5]) cots) = PCx E51) - PCN ES2) Ys) eo) ethm- Tf X44 ave two independent He “Fandom variable € ge-) ¢ hE) are. bored Ip xa are Fanchions then gca) & h(Y) are alse independent yndependenk of each other- Fra (oy) = Fx FY) an A) =—4 Kaa Oe independent ous o(K)& oY) are independent 2303), | of each other \ Farther sC9CK)) SeCK) & oChiy)) Cty) -0 ie Areccgcey) & Aa €eChyy)) —@® Tok independent of each other C9) & oChty)) are ata) €e4) are independent oF each other~ praoms=PLAy. PLA] \ Bina) e o(9cu¢ oChly) indepk- (086%) are J are indep | independent | oF each (simi lart “TF x i 9 degenorate “OV then X is ) ae tanta) independent every other mu, ve) = Kei) ore = €K0CsD /sen} = Lad can! be any other mn Inverse. image ; let AEscy) of degenerate fa . vis tiuld | ™ previous svesult, every event Rela { is independent of o ¢ 4 ele) & et) are independent oF each other: “xX AN are independent. © vegenerate. au is independent of itself - let X be A degenerote mv otk) = (4,23 ¢ is independent of sis independent ot ¢ xis independent of itself © TE X &¥ ave indepb mus then EGY= EOD EW) ECKY) = EUS-EUY) — caset] x&Y both are simple mv xy dre independent K = Jailai a Ee ‘ 3 ctx) & CN) are independent qs 5 bj Taj oF each other a: eO)= {Bn.-- Alk} Unien of mic Kx ; x} =2 Z aibj Imag, OCH) = [Bi 82,..81,-_ 958} eet ne 3 every Ai ts independent 3 xy -is simple TY oF evry By ey, . \ ; ECKY) = BZ aikj FCaing) EON) =, = & aiej PCH): 2685) jalget \S . om 3 arco |[3 nrta| = EQ9-ECY) 3 ECxY) = EOO-ECY) case TI} 2 X & both ate non negative w-vs = EEX) = lim ECA) where Can a2} & (Yai net} are ~e pon decreasing sequence of ECN) = a EC%n) non-negative simple wu E (KY) =lim EC¥nn) Knt% > Yn} me anna KY, Kn SX} = Nn ECan} ECNo) Nn £4nn Jun v =e Framiqse Xn S$ Xnt} = ECO: EC) yon SXmn Yn Yay yng No, xny do & \nel Xn =) yan SNankaty

You might also like