Lecture #17: Expectation of a Simple Random Variable

Size: px
Start display at page:

Download "Lecture #17: Expectation of a Simple Random Variable"

Transcription

1 Statistics 851 (Fall 2013) October 16, 2013 Prof. Michael Kozdro Lecture #17: Expectatio of a Simple Radom Variable Recall that a simple radom variable is oe that takes o fiitely may values. Defiitio. Let (Ω, F, P) beaprobabilityspace. AradomvariableX :Ω R is called simple if it ca be writte as X = a i 1 Ai where a i R, A i F for i =1, 2,...,.WedefietheexpectatioofX to be E(X) = a i P {A i }. Example Cosider the probability space (Ω, B 1, P) whereω=[0, 1], B 1 deotes the Borel sets of [0, 1], ad P is the uiform probability o Ω. Suppose that the radom variable X :Ω R is defied by 4 X(ω) = a i 1 Ai (ω) where a 1 =4,a 2 =2,a 3 =1,a 4 = 1, ad A 1 =[0, 1 2 ), A 2 =[ 1 4, 3 4 ), A 3 =( 1 2, 7 8 ], A 4 =[ 7 8, 1]. Show that there exist fiitely may real costats c 1,...,c ad disjoit sets C 1,...,C B 1 such that X = c i 1 Ci. Solutio. We fid so that 4, if 0 ω<1/4, 6, if 1/4 ω<1/2, 2, if ω =1/2, X(ω) = 3, if 1/2 <ω<3/4, 1, if 3/4 ω<7/8, 0, if ω =7/8, 1, if 7/8 <ω 1, X = 7 c i 1 Ci where c 1 =4,c 2 =6,c 3 =2,c 4 =3,c 5 =1,c 6 =0,c 7 = 1 ad C 1 =[0, 1 4 ), C 2 =[ 1 4, 1 2 ), C 3 = { 1 2 }, C 4 =( 1 2, 3 4 ), C 5 =[ 3 4, 7 8 ), C 6 = { 7 8 }, C 7 =( 7 8, 1]. 17 1

2 Propositio If X ad Y are simple radom variables, the E(αX + βy )= αe(x)+ βe(y ) for every α, β R. Proof. Suppose that X ad Y are simple radom variables with X = a i 1 Ai ad Y = m b j 1 Bj where A 1,...,A F ad B 1,...,B m F each partitio Ω. Sice αx = α a i 1 Ai = (αa i )1 Ai we coclude by defiitio that E(αX) = (αa i )P {A i } = α a i P {A i } = αe(x). The proof of the theorem will be completed by showig E(X + Y )=E(X)+E(Y ). Notice that {A i B j :1 i, 1 j m} cosists of pairwise disjoit evets whose uio is Ω ad Therefore, by defiitio, X + Y = m (a i + b j )1 Ai B j. E(X + Y )= = = m (a i + b j )P {A i B j } m a i P {A i B j } + a i P {A i } + m b j P {A i B j } m b j P {B j } ad the proof is complete. Fact. If X ad Y are simple radom variables with X Y,the E(X) E(Y ). Exercise Prove the previous fact. 17 2

3 Havig already defied E(X) for simple radom variables,our goal ow is to costruct E(X) i geeral. To that ed, suppose that X is a positive radom variable. That is, X(ω) 0 for all ω Ω. (We will eed to allow X(ω) [0, +] forsomecosistecy.) Defiitio. If X is a positive radom variable, defie the expectatio of X to be E(X) = sup{e(y ):Y is simple ad 0 Y X}. That is, we approximate positive radom variables by simple radom variables. Of course, this leads to the questio of whether or ot this is possible. Fact. For every radom variable X 0, there exists a sequece (X )ofpositive,simple radom variables with X X (that is, X icreases to X). A example of such a sequece is give by k k, if X(ω) < k+1 ad 0 k 2 1, 2 X (ω) = 2 2, if X(ω). (Draw a picture.) Fact. If X 0ad(X )isasequeceofsimpleradomvariableswithx E(X ) E(X). X, the We will prove these facts ext lecture. Now suppose that X is ay radom variable. Write X + =max{x, 0} ad X = mi{x, 0} for the positive part ad the egative part of X, respectively. Note that X + 0adX 0sothatthepositivepartadegativepartofX are both positive radom variables ad X = X + X ad X = X + + X. Defiitio. AradomvariableX is called itegrable (or has fiite expectatio) ifboth E(X + )ade(x )arefiite.ithiscasewedefiee(x) tobe E(X) =E(X + ) E(X ). 17 3

4 Statistics 851 (Fall 2013) October 18, 2013 Prof. Michael Kozdro Lecture #18: Costructio of Expectatio Recall that our goal is to defie E(X) for all radom variables X :Ω R. Weoutliedthe costructio last lecture. Here is the summary of our strategy. Summary of Strategy for Costructig E(X) for Geeral Radom Variables We will (1) defie E(X) for simple radom variables, (2) defie E(X) for positive radom variables, (3) defie E(X) for geeral radom variables. This strategy is sometimes called the stadard machie ad is the outlie that we will follow to prove most results about expectatio of radom variables. Step 1: Simple Radom Variables Let (Ω, F, P) beaprobabilityspace. SupposethatX :Ω R is a simple radom variable so that m X(ω) = a j 1 Aj (ω) where a 1,...,a m R ad A 1,...,A m F.WedefietheexpectatioofX to be E(X) = m a j P {A j }. Step 2: Positive Radom Variables Suppose that X is a positive radom variable. That is, X(ω) 0forallω Ω. (We will eed to allow X(ω) [0, +] forsomecosistecy.)wearealsoassumigatthisstepthat X is ot a simple radom variable. Defiitio. If X is a positive radom variable, defie the expectatio of X to be E(X) = sup{e(y ):Y is simple ad 0 Y X}. (18.1) Propositio For every radom variable X 0, there exists a sequece (X ) of positive, simple radom variables with X X (that is, X icreases to X). 18 1

5 Proof. Let X 0begiveaddefiethesequece(X )by k k, if X(ω) < k+1 ad 0 k 2 1, 2 X (ω) = 2 2, if X(ω). The it follows that X X +1 for every =1, 2, 3,... ad X X which completes the proof. Propositio If X 0 ad (X ) is a sequece of simple radom variables with X X, the E(X ) E(X). That is, if X X +1 ad lim X = X, the E(X +1 ) E(X ) ad lim E(X )=E(X). Proof. Suppose that X 0isaradomvariableadlet(X )beasequeceofsimple radom variables with X 0adX X. Observe that sice the X are icreasig, we have E(X ) E(X +1 ). Therefore, E(X )icreasestosomelimita [0, ]; that is, E(X ) a for some 0 a.(ife(x )isauboudedsequece,thea =. However, if E(X ) is a bouded sequece, the a< follows from the fact that icreasig, bouded sequeces have uique limits.) Therefore, it follows from (18.1), the defiitio of E(X), that a E(X). We will ow show a E(X). As a result of (18.1), we oly eed to show that if Y is a simple radom variable with 0 Y X, thee(y ) a. That is, by defiitio, E(X) = sup{e(y ):Y is simple with 0 Y X} ad so if Y is a arbitrary simple radom variable satisfyig 0 Y X ad E(Y ) a, thethedefiitioofsupremumimpliese(x) a. To this ed, let Y be simple ad write Y = m a k 1{Y = a k }. That is, take A k = {ω : Y (ω) =a k }.Let0< 1addefie Y, =(1 )Y1 {(1 )Y X}. Note that Y, =(1 )a k o the set {(1 )Y X } A k = A k,, ad that Y, =0otheset{(1 )Y >X }.ClearlyY, X ad so m E(Y, )=(1 ) a k P {A k,, } E(X ). 18 2

6 We will ow show that A k,, icreases to A k. Sice X X +1 ad X X we coclude that {(1 )Y X } {(1 )Y X +1 } {(1 )Y X} ad therefore A k,, A k,+1, {(1 )Y X} A k. (18.2) Sice Y X by assumptio, we kow that the evet {(1 )Y X} is equal to Ω. Thus, (18.2) implies A k,, A k for all ad so A k,, A k. (18.3) =1 Coversely, let ω A k = {(1 )Y X} A k = {ω :(1 )Y (ω) X(ω)} A k so that (1 )a k X(ω). Sice Y X, whichistosaythat Y (ω) X(ω) = lim X (ω) for all ω, wekowthatifω A k,thethereissomen such that Y (ω) =a k X (ω) wheever >N.Wethereforecocludethatifω A k ad >N,the(1 )a k <X (ω). (This requires that X is ot idetically 0.) That is, ω A k,, which proves that A k,, A k. (18.4) =1 I other words, it follows from (18.3) ad (18.4) that A k,, icreases to A k, so by the cotiuity of probability theorem (Theorem 10.2), we coclude that Hece, That is, ad so that E(Y, )=(1 ) lim P {A k,,} = P {A k }. m a k P {A k,, } (1 ) m a k P {A k } =(1 )E(Y ) a. E(Y, ) E(X ) E(Y, ) (1 )E(Y ) ad E(X ) a (1 )E(Y ) a (which follows sice everythig is icreasig). Sice 0 < 1isarbitrary, E(Y ) a which, as oted earlier i the proof, is sufficiet to coclude that E(X) a. Combied with our earlier result that a E(X) wecocludee(x) =a ad the proof is complete. 18 3

7 Step 3: Geeral Radom Variables Now suppose that X is ay radom variable. Write X + =max{x, 0} ad X = mi{x, 0} for the positive part ad the egative part of X, respectively. Note that X + 0ad X 0sothatthepositivepartadegativepartofX are both positive radom variables ad X = X + X. Defiitio. AradomvariableX is called itegrable (or has fiite expectatio) ifboth E(X + )ade(x )arefiite.ithiscasewedefiee(x) tobe E(X) =E(X + ) E(X ). Defiitio. If oe of E(X + )ore(x )isifiite,thewecastilldefiee(x) by settig +, if E(X + )=+ ad E(X ) <, E(X) =, if E(X + ) < ad E(X )=+. However, X is ot itegrable i this case. Defiitio. If both E(X + )=+ ad E(X )=+, thee(x) doesotexist. Remark. We see that the stadard machie is really ot that hard to implemet. I fact, it is usually eough to prove a result for simple radom variables ad the exted that result to positive radom variables usig Propositios 18.1 ad The result for geeral radom variables usually the follows by defiitio. 18 4

8 Statistics 851 (Fall 2013) October 21, 2013 Prof. Michael Kozdro Lecture #19: Expectatio ad Itegratio Defiitio. Let (Ω, F, P) beaprobabilityspace,adletx :Ω R be a radom variable. If X is a simple radom variable, say X(ω) = a i 1 Ai (ω) for a 1,...,a R ad A 1,...,A F,thewedefie E(X) = a i P {A i }. If X is a positive radom variable, the we defie E(X) = sup{e(y ):Y is simple ad 0 Y X}. If X is ay radom variable, the we ca write X = X + X where both X + ad X are positive radom variables. Provided that both E(X + )ade(x )arefiite,wedefiee(x) to be E(X) =E(X + ) E(X ) ad we say that X is itegrable (or has fiite expectatio). Remark. If X :(Ω, F, P) (R, B) is a radom variable, the we sometimes write E(X) = X dp = X(ω)dP {ω} = X(ω)P {dω}. Ω Ω That is, the expectatio of a radom variable is the Lebesgue itegral of X. We will say more about this later. Defiitio. Let L 1 (Ω, F, P) bethesetofreal-valuedradomvariableso(ω, F, P) with fiite expectatio. That is, L 1 (Ω, F, P) ={X :(Ω, F, P) (R, B) : X is a radom variable with E(X) < }. Ω We will ofte write L 1 probability space. for L 1 (Ω, F, P) adsuppressthedepedeceotheuderlyig Theorem Suppose that (Ω, F, P) is a probability space, ad let X 1,X 2,..., X, ad Y all be real-valued radom variables o (Ω, F, P). (a) L 1 is a vector space ad expectatio is a liear map o L 1. Furthermore, expectatio is positive. That is, if X, Y L 1 with 0 X Y, the 0 E(X) E(Y ). 19 1

9 (b) X L 1 if ad oly if X L 1, i which case we have E(X) E( X ). (c) If X = Y almost surely (i.e., if P {ω : X(ω) =Y (ω)} =1), the E(X) =E(Y ). (d) (Mootoe Covergece Theorem) If the radom variables X 0 for all ad X X (i.e., X X ad X X +1 ), the lim E(X )=E lim X = E(X). (We allow E(X) =+ if ecessary.) (e) (Fatou s Lemma) If the radom variables X all satisfy X Y almost surely for some Y L 1 ad for all, the E lim if X lim if E(X ). (19.1) I particular, (19.1) holds if X 0 for all. (f) (Lebesgue s Domiated Covergece Theorem) If the radom variables X X, ad if for some Y L 1 we have X Y almost surely for all, the X L 1, X L 1, ad lim E(X )=E(X). Remark. This theorem cotais ALL of the cetral results of Lebesgue itegratio theory. Theorem Let X be a sequece of radom variables. (a) If X 0 for all, the (b) If E X = =1 E(X ) (19.2) =1 with both sides simultaeously beig either fiite or ifiite. the E( X ) <, =1 coverges almost surely to some radom variable Y L 1. I other words, =1 = X X

10 is itegrable with E X = =1 Thus, (19.2) holds with both sides beig fiite. E(X ). Notatio. For 1 p<, letl p = {radom variables X :Ω R such that X p L 1 }. Theorem 19.3 (Cauchy-Schwartz Iequality). If X, Y L 2, the XY L 1 ad =1 E(XY ) E(X 2 )E(Y 2 ). Theorem Let X :(Ω, F, P) (R, B) be a radom variable. (a) (Markov s Iequality) If X L 1, the for every a>0. (b) (Chebychev s Iequality) If X L 2, the for every a>0. P { X a} E( X ) a P { X a} E(X2 ) a

11 Statistics 851 (Fall 2013) October 23, 2013 Prof. Michael Kozdro Lecture #20: Proofs of the Mai Expectatio Theorems Our goal for today is to start provig all of the importat results for expectatio that were stated last lecture. Note that the proofs geerally follow the so-called stadard machie; that is, we first prove the result for simple radom variables, the exted it to o-egative radom variables, ad fially exted it to geeral radom variables. The key results for implemetig this strategy are Propositio 18.1 ad Propositio Theorem Let (Ω, F, P) be a probability space. If L 1 deotes the space of itegrable radom variables, amely L 1 = {radom variables X such that E(X) < }, the L 1 is a vector space ad expectatio is a liear operator o L 1. Moreover, expectatio is mootoe i the sese that if X ad Y are radom variables with 0 X Y ad Y L 1, the X L 1 ad 0 E(X) E(Y ). Proof. Suppose that X 0adY 0areo-egativeradomvariables,adletα [0, ). We kow that there exist sequeces X ad Y of o-egative simple radom variables such that (i) X X ad E(X ) E(X), ad (ii) Y Y ad E(Y ) E(Y ). Therefore, αx is a sequece of o-egative simple radom variables with (αx ) (αx) ad E(αX ) E(αX ). Moreover, X + Y is a sequece of o-egative simple radom variables with (X + Y ) (X + Y )ade(x + Y ) E(X + Y ). However, we kow that expectatio is liear o simple radom variables so that Thus, we fid E(αX )=αe(x ) ad E(X + Y )=E(X )+E(Y ). E(αX ) = αe(x ) E(αX) αe(x) ad E(X + Y ) = E(X )+E(Y ) E(X + Y ) E(X)+E(Y ) so by uiqueess of limits, we coclude E(αX) =αe(x) ade(x + Y )=E(X + Y ). Also ote that if 0 X Y,thethedefiitioofexpectatioofo-egativeradomvariables immediately implies that 0 E(X) E(Y )sothaty L 1 implies X L 1. Suppose ow that X ad Y are geeral radom variables ad let α R. Sice ad (αx) =(αx) + (αx) (αx) + +(αx) α (X + + X ) (X + Y )=(X + Y ) + (X + Y ) (X + Y ) + +(X + Y ) X + + X + Y + + Y. Hece, sice X +, X, Y +, Y L 1,wecocludethatαX L 1 ad X + Y L 1. Fially, sice E(X) =E(X + ) E(X )bydefiitio,adsiceweshowedabovethatexpectatiois liear o o-egative radom variables, we coclude that expectatio is liear o geeral radom variables. 20 1

12 Theorem If X :(Ω, F, P) (R, B) is a radom variable, the X L 1 if ad oly if X L 1. Proof. Suppose that X L 1 so that E(X + ) < ad E(X ) <. Sice X = X + + X ad sice expectatio is liear, we coclude that so that X L 1. E( X ) =E(X + + X )=E(X + )+E(X ) < O the other had, suppose that X L 1 so that E(X + )+E(X ) <. However, sice X + 0adX 0, we kow that E(X + ) 0adE(X ) 0. We ow use the fact that if the sum of two o-egative umbers is fiite, the each umber must be fiite to coclude that E(X + ) < ad E(X ) <. Thus, by defiitio, E(X) =E(X + ) E(X ) < so that X L 1. Corollary. If X L 1, the E(X) E( X ). I particular, if E( X ) =0, the E(X) =0. Proof. Sice E(X + ) 0adE(X ) 0, we have from the triagle iequality E(X) = E(X + X ) = E(X + ) E(X ) E(X + ) + E(X ) = E(X + )+E(X )=E( X ). Sice E(X) 0, if E( X ) =0,the0 E(X) E( X ) =0implyigE(X) =0. Theorem If X, Y :(Ω, F, P) (R, B) are itegrable radom variables with X = Y almost surely, the E(X) =E(Y ). Proof. Suppose that X = Y almost surely so that P {ω Ω:X(ω) =Y (ω)} =1. Tobegi, assume that X 0adY 0adletA = {ω : X(ω) = Y (ω)} so that P {A} =0. Write E(Y )=E(Y 1 A + Y 1 A c)=e(y 1 A )+E(Y 1 A c)=e(y 1 A )+E(X1 A c). ( ) We kow that there exist sequeces X ad Y of o-egative simple radom variables such that (i) X X ad E(X ) E(X), ad (ii) Y Y ad E(Y ) E(Y ). Thus, X 1 A X1 A ad E(X 1 A ) E(X1 A )adsimilarlyy 1 A Y 1 A ad E(Y 1 A ) E(Y 1 A ). For each, the radom variable X takes o fiitely may values ad is therefore bouded by K, say,where K may deped o. Thus,siceX K, we obtai X 1 A K ad so 0 E(X 1 A ) E(K1 A )=KP {A} =0. This implies that E(X 1 A )=0adsobyuiqueessoflimits,E(X1 A )=0. E(Y 1 A )=0. Therefore,by( ), we obtai Similarly, E(Y )=E(Y 1 A )+E(X1 A c)=0+e(x1 A c)=e(x1 A )+E(X1 A c)=e(x). I geeral, ote that X = Y almost surely implies that X + = Y + almost surely ad X = Y almost surely. 20 2

13 Statistics 851 (Fall 2013) October 25, 2013 Prof. Michael Kozdro Lecture #21: Proofs of the Mai Expectatio Theorems (cotiued) We will cotiue provig the importat results for expectatio that were stated i Lecture #19. Recall that a radom variable is said to have fiite mea or have fiite expectatio or be itegrable if E(X) <. Thevectorspaceofallitegrableradomvariableoagive probability space (Ω, F,P)isdeotedbyL 1 ad if 1 p<, the L p = {radom variables X :Ω R such that X p L 1 }. Theorem 21.1 (Cauchy-Schwartz Iequality). If X, Y L 2, the XY L 1 ad E(XY ) E(X 2 )E(Y 2 ). Proof. Sice 0 (X + Y ) 2 = X 2 + Y 2 +2XY ad 0 (X Y ) 2 = X 2 + Y 2 2XY,we coclude that 2 XY X 2 + Y 2 implyig 2E( XY ) E(X 2 )+E(Y 2 ). Thus, if X, Y L 2, we coclude that XY L 1.Foreveryx R, otethat 0 E((xX + Y ) 2 )=x 2 E(X 2 )+2xE(XY )+E(Y 2 ). Sice x 2 E(X 2 )+2xE(XY )+E(Y 2 )isao-egativequadraticix, itsdiscrimiatis ecessarily o-positive; that is, 4[E(XY )] 2 4E(X 2 )E(Y 2 ) 0, or, equivaletly, as required. E(XY ) E(X 2 )E(Y 2 ) Theorem Let X :(Ω, F, P) (R, B) be a radom variable. (a) (Markov s Iequality) If X L 1, the for every a>0. P { X a} E( X ) a (b) (Chebychev s Iequality) If X L 2, the for every a>0. P { X a} E(X2 ) a

14 Proof. Recall that X L 1 if ad oly if X L 1.Sice X 0, we ca write X a1 { X a}. Takig expectatios of the previous expressio implies E( X ) ae(1 { X a} )=P { X a} ad Markov s iequality follows. Similarly, X 2 a 2 1 {X 2 a 2 } so that if X L 2,takig expectatios implies E(X 2 ) a 2 P X 2 a 2. Hece, P { X a} P X 2 a E(X2 ) a 2 yieldig Chebychev s iequality. Defiitio. If X L 2 we say that X has fiite variace ad defie the variace of X to be Var(X) =E[(X E(X)] 2 )=E(X 2 ) [E(X)] 2. Thus, Chebychev s iequality sometimes takes the form Computig Expectatios P { X E(X) a} Var(X) a 2. Havig proved a umber of the mai expectatio theorems, we will ow take a small detour to discuss computig expectatios. We will also show over the course of the ext several lectures how the formulas for the expectatios of discrete ad cotiuous radom variables ecoutered i itroductory probability follow from the geeral theory developed. The first examples we will discuss, however, are the calculatios of expectatios directly from the defiitio ad theory. Example Cosider ([0, 1], B 1, P) whereb 1 are the Borel sets of [0, 1] ad P is the uiform probability. Let X :Ω R be give by X(ω) =ω so that X is a uiform radom variable. We will ow compute E(X) directly by defiitio. Note that X is ot simple sice the rage of X, amely[0, 1], is ucoutable. However, X is positive. This meas that as acosequeceofpropositios18.1ad18.2,ifx is a sequece of positive, simple radom variables with X X, thee(x ) E(X). Thus, let 0, 0 ω<1/2, X 1 (ω) = 1/2, 1/2 ω 1 0, 0 ω 1/4, 1/4, 1/4 ω<1/2, X 2 (ω) = 1/2, 1/2 ω<3/4, 3/4, 3/4 ω 1, 21 2

15 ad i geeral, X (ω) = 2 2 j 1 j ω< j ω Thus, implyig E(X )= = j 1 j 1 2 P 2, j P, j 1 2 = (j 1) 2 2 j 2 j 1 2 = 1 (2 2)(2 1) = E(X) = lim E(X )= lim = Example Cosider ([0, 1], B 1, P) whereb 1 are the Borel sets of [0, 1] ad P is the uiform probability. Let Q 1 =[0, 1] Q ad cosider the radom variable Y :Ω R give by ω, if ω [0, 1] \ Q 1, Y (ω) = 0, if ω Q 1. Note that Y is ot a simple radom variable, although it is o-egative. I order to compute E(Y )itiseasiesttousetheorem20.3.thatis,letx(ω) =ω for ω [0, 1] ad ote that {ω : X(ω) = Y (ω)} = Q 1 \{0}. (The techical poit here is that X(0) = Y (0) = 0. However, if ω Q 1 with ω = 0,the X(ω) = Y (ω).) Sice P {Q 1 \{0}} =0,weseethatX ad Y differ o a set of probability 0 so that X = Y almost surely. Sice X is a uiform radom variable o [0, 1], we kow from the previous example that E(X) = 1/2. Therefore, usig Theorem 20.3 we coclude E(Y )=E(X) =1/

16 Statistics 851 (Fall 2013) October 28, 2013 Prof. Michael Kozdro Lecture #22: Computig Expectatios of Discrete Radom Variables Recall from itroductory probability classes that a radom variable X is called discrete if the rage of X is at most coutable. The formula for the expectatio of a discrete radom variable X give i these classes is E(X) = jp {X = j}. We will ow derive this formula as a cosequece of the geeral theory developed durig the past several lectures. Suppose that the rage of X is at most coutable. Without loss of geerality, we ca assume that Ω = {1, 2,...,}, F =2 Ω,adX :Ω R is give by X(ω) =ω. Assume further that the law of X is give by P {X = j} = p j where p j [0, 1] ad p j =1. Let A j = {X = j} = {ω : X(ω) =j} so that X ca be writte as X(ω) = j1 Aj. Observe that if p j =0forifiitelymayj, thex is simple ad ca be writte as X(ω) = j 1 Aj for some < which implies that E(X) = j P {A j } = jp {X = j}. O the other had, suppose that p j =0forifiitelymayj. Ithiscase,X is ot simple. We ca approximate X by simple fuctios as follows. Let A j, = {ω : X(ω) =j, j } so that A j, A j,+1 ad A j, = A j. =1 22 1

17 That is, A j, A j ad so by cotiuity of probability we coclude If we ow set so that X (ω) = lim P {A j,} = P {A j }. X(ω)1 Aj, = E(X )= jp {A j, }, j1 Aj, the (X )isasequeceofpositivesimpleradomvariablewithx X. Propositio 18.2 the implies as required. E(X) = lim E(X )= lim jp {A j, } = P {A j } = jp {X = j} Computig Expectatios of Cotiuous Radom Variables We will ow tur to that other formula for computig expectatio ecoutered i elemetary probability courses, amely if X is a cotiuous radom variable with desity f X,the E(X) = xf X (x)dx. It turs out that verifyig this formula is somewhat more ivolved tha the discrete formula. As such, we eed to take a brief detour ito some geeral fuctio theory. Some Geeral Fuctio Theory Suppose that f : X Y is a fuctio. We are implicitly assumig that f is defied for all x X. WecallX the domai of f ad call Y the codomai of f. The rage of f is the set f(x) ={y Y : f(x) =y for some x X}. Note that f(x) Y. Iff(X) =Y, thewesaythatf is oto Y. Let B Y. Wedefief 1 (B) by f 1 (B) ={x X : f(x) =y for some y B} = {f B} = {x : f(x) B}. We call X a topological space if there is a otio of ope subsets of X. TheBorelσ-algebra o X, writteb(x), is the σ-algebra geerated by the ope sets of X. 22 2

18 Let X ad Y be topological spaces. A fuctio f : X Y is called cotiuous if for every ope set V Y, thesetu = f 1 (V ) X is ope. Afuctiof : X Y is called measurable if f 1 (B) B(X) foreveryb B(Y). Sice the ope sets geerate the Borel sets, this theorem follows easily. Theorem Suppose that (X, B(X)) ad (Y, B(Y)) are topological measure spaces. The fuctio f : X Y is measurable if ad oly if f 1 (O) B(X) for every ope set O B(Y). The ext theorem tells us that cotiuous fuctios are ecessarily measurable fuctios. Theorem Suppose that (X, B(X)) ad (Y, B(Y)) are topological measure spaces. If f : X Y is cotiuous, the f is measurable. Proof. By defiitio, f : X Y is cotiuous if ad oly if f 1 (O) X is a ope set for every ope set O Y. Sice a ope set is ecessarily a Borel set, we coclude that f 1 (O) B(X) foreveryopeseto B(Y). However, it ow follows immediately from the previous theorem that f : X Y is measurable. The followig theorem shows that the compositio of measurable fuctios is measurable. Theorem Suppose that (W, F), (X, G), ad (Y, H) are measurable spaces, ad let f :(W, F) (X, G) ad g :(X, G) (Y, H) be measurable. The the fuctio h = g f is a measurable fuctio from (W, F) to (Y, H). Proof. Suppose that H H. Sice g is measurable, we have g 1 (H) G. Sice f is measurable, we have f 1 (g 1 (H)) F.Sice the proof is complete. h 1 (H) =(g f) 1 (H) =f 1 (g 1 (H)) F 22 3

19 Statistics 851 (Fall 2013) October 30, 2013 Prof. Michael Kozdro Lecture #23: Proofs of the Mai Expectatio Theorems (cotiued) Theorem 23.1 (Mootoe Covergece Theorem). Suppose that (Ω, F, P) is a probability space, ad let X 1,X 2,..., ad X be real-valued radom variables o (Ω, F, P). If X 0 for all ad X X (i.e., X X ad X X +1 ), the lim E(X )=E (allowig E(X) =+ if ecessary). lim X = E(X) Proof. For every, lety,k, k =1, 2,..., be o-egative ad simple with Y,k X ad E(Y,k ) E(X )ask which is possible by Propositios 18.1 ad Set Z k =max k Y,k, ad observe that 0 Z k Z k+1 so that (Z k )isaicreasigsequeceofo-egative simple radom variables which ecessarily has a limit Z =lim k Z k. By Propositio 18.2, E(Z k ) E(Z). We ow observe that if k, the Y,k Z k X X k X. (23.1) We ow deduce from (23.1) that X Z X almost surely, ad so lettig implies X = Z almost surely. We also deduce from (23.1) that for k. Fix ad let k to obtai E(Y,k ) E(Z k ) E(X k ) Now let to obtai E(X ) E(Z) lim k E(X k ). Thus, lim E(X ) E(Z) lim E(X k ). k E(Z) = lim E(X ). But X = Z almost surely so that E(X) =E(Z) adwecoclude as required. E(X) = lim E(X ) 23 1

20 Theorem 23.2 (Fatou s Lemma). Suppose that (Ω, F, P) is a probability space, ad let X 1,X 2,... be real-valued radom variables o (Ω, F, P). IfX Y almost surely for some Y L 1 ad for all, the E lim if X lim if E(X ). (23.2) I particular, (23.2) holds if X 0 for all. Proof. Without loss of geerality, we ca assume X 0. Here is the reaso. If X = X Y, the X L 1 with X 0ad E lim if X lim if E( X ) if ad oly if E lim if X lim if E(X ) because Hece, if X 0, set lim if X = lim if X Y. Y =if k X k so that Y 0isaradomvariableadY Y +1.ThisimpliesthatY coverges ad lim Y =limif X. Sice X Y,mootoicityofexpectatioimpliesE(X ) E(Y ). This yields lim if E(X ) lim E(Y )=E lim Y =limif E(X ) by the Mootoe Covergece Theorem ad the proof is complete. Theorem 23.3 (Lebesgue s Domiated Covergece Theorem). Suppose that (Ω, F, P) is a probability space, ad let X 1,X 2,... be real-valued radom variables o (Ω, F, P). If the radom variables X X, ad if for some Y L 1 we have X Y almost surely for all, the X L 1, X L 1, ad lim E(X )=E(X). Proof. Suppose that we defie the radom variables U ad V by U =limif X ad V =limsupx. The assumptio that X X implies that U = V = X almost surely so that E(U) = E(V )=E(X). Ad if we also assume that X Y almost surely for all so that X L 1, the we coclude X Y. Sice Y L 1,wehaveX L 1. Moreover, X Y almost surely ad Y L 1 so by Fatou s Lemma, E(U) lim if E(X ). 23 2

21 However, we also kow X Y almost surely ad so Fatou s lemma also implies V =limif ( X ) E(V )=E( V ) lim if Combiig these two iequalities gives E(X) =E(U) lim if so that E(X ) E(X) asrequired. E( X )= lim sup E(X ). E(X ) lim sup E(X ) E(V )=E(X) Theorem Let X be a sequece of radom variables. (a) If X 0 for all, the (b) If E X = =1 E(X ) (23.3) =1 with both sides simultaeously beig either fiite or ifiite. the E( X ) <, =1 coverges almost surely to some radom variable Y L 1. I other words, is itegrable with =1 =1 X X E X = =1 Thus, (23.3) holds with both sides beig fiite. Proof. Suppose that S = E(X ). =1 X k ad T = X k. Sice S cotais fiitely may terms, we kow by liearity of expectatio that E(S )=E X k = E( X k ). 23 3

22 Moreover, 0 S S +1 so that S icreases to some limit S = X k. Note that S(ω) [0, +]. By the Mootoe Covergece Theorem, E(S) = lim E(S )= lim E( X k )= E( X k ). If X 0forall, thes = T ad (a) follows. If X are geeral radom variables, with E( X k ) <, the E(S) <. Now, for every > 0, sice 1 S= S, wecocludep {S = } = E(1 S= ) E(S). Sice >0is arbitrary ad E(S) <, wecocludethatp {S = } =0. Therefore, =1 is absolutely coverget almost surely ad its sum is the limit of the sequece T.Moreover, T S S ad S L 1,sobytheDomiatedCovergeceTheorem, E X k = E lim X k = lim E(X k )= E(X k ) provig (b). X 23 4

23 Statistics 851 (Fall 2013) November 1, 2013 Prof. Michael Kozdro Lecture #24: The Expectatio Rule Let (Ω, F, P) beaprobabilityspace,adletx :Ω R be a radom variable so that X 1 (B) Ffor every B B. Suppose that h : R R is a measurable fuctio so that h 1 (B) Bfor every B B. Propositio The fuctio h X :Ω R give by h X(ω) =h(x(ω)) is a radom variable. Proof. To prove that h X is a radom variable, we must show that (h X) 1 (B) = X 1 (h 1 (B)) Ffor every B B. Thus, suppose that B B. Sice h is measurable, we kow h 1 (B) B. Let A = h 1 (B). Sice A is a Borel set, i.e., A B,wekowthat X 1 (A) F sice X is a radom variable. That is, (h X) 1 (B) =X 1 (h 1 (B)) = X 1 (A) F so that h X :Ω R is a radom variable as required. We kow that X iduces a probability o (R, B) calledthelaw of X ad defied by P X {B} = P X 1 (B) = P {X B} = P {ω Ω:X(ω) B}. I other words, (R, B, P X )isaprobabilityspace. Thismeasthatifh : R R is a measurable fuctio o (R, B, P X ), the we ca also call h a radom variable. As such, it is possible to discuss the expectatio of h, amely E(h) = h(x) P X {dx}. R As we will ow show, the expectatio of h ad the expectatio of h X are itimately related. Theorem 24.2 (Expectatio Rule). Let X :(Ω, F, P) (R, B) be a radom variable with law P X, ad let h :(R, B) (R, B) be measurable. (a) The radom variable h(x) L 1 (Ω, F, P) if ad oly if h L 1 (R, B, P X ). (b) If h 0, or if either coditio i (a) holds, the E(h(X)) = h(x(ω)) P {dω} = Ω R h(x) P X {dx}. Remark. The secod itegral i the theorem is actually a Lebesgue itegral sice it is the itegral of a radom variable, amely h, with respect to a probability, amely P X. As we will see shortly, we ca always relate this to a Riema-Stieltjes itegral ad i some cases to a Riema itegral. 24 1

24 Proof. Sice h X :Ω R is a radom variable by Propositio 24.1, we kow E(h(X)) = h(x(ω)) P {dω} by defiitio. The cotet of the theorem is that this itegral is equal to h(x) P X {dx}. R Ω I order to complete the proof, we will follow the stadard machie. That is, let h(x) =1 B (x) for B Bso that E(h(X)) = E(1 B (X)) = P {X B} = P X 1 (B) = P X {B} = P X {dx} B = 1 B (x) P X {dx} R = h(x) P X {dx}. Hece, if h is simple, say h(x) = b i 1 Bi (x) for some b 1,...,b R ad B 1,...,B B,the E(h(X)) = E b i 1 Bi (x) = b i P X {B i } = = = R R b i b i 1 Bi (x) P X {dx} h(x) P X {dx} R R 1 Bi (x) P X {dx} usig liearity of expectatio. If h 0, let h 0besimplewithh h so that E(h(X)) = E lim h (X) = lim E(h (X)) = lim h (x) P R X {dx} = lim h (x) P X {dx} R = h(x) P X {dx} usig the Mootoe Covergece Theorem twice. I particular, (b) follows for h 0. If we apply the above procedure to h, the(a)follows. Ifh is ot positive, the writig h = h + h implies the result by subtractio. R 24 2

Convergence of random variables. (telegram style notes) P.J.C. Spreij

Convergence of random variables. (telegram style notes) P.J.C. Spreij Covergece of radom variables (telegram style otes).j.c. Spreij this versio: September 6, 2005 Itroductio As we kow, radom variables are by defiitio measurable fuctios o some uderlyig measurable space

More information

Lecture 3 : Random variables and their distributions

Lecture 3 : Random variables and their distributions Lecture 3 : Radom variables ad their distributios 3.1 Radom variables Let (Ω, F) ad (S, S) be two measurable spaces. A map X : Ω S is measurable or a radom variable (deoted r.v.) if X 1 (A) {ω : X(ω) A}

More information

Lecture 3 The Lebesgue Integral

Lecture 3 The Lebesgue Integral Lecture 3: The Lebesgue Itegral 1 of 14 Course: Theory of Probability I Term: Fall 2013 Istructor: Gorda Zitkovic Lecture 3 The Lebesgue Itegral The costructio of the itegral Uless expressly specified

More information

Product measures, Tonelli s and Fubini s theorems For use in MAT3400/4400, autumn 2014 Nadia S. Larsen. Version of 13 October 2014.

Product measures, Tonelli s and Fubini s theorems For use in MAT3400/4400, autumn 2014 Nadia S. Larsen. Version of 13 October 2014. Product measures, Toelli s ad Fubii s theorems For use i MAT3400/4400, autum 2014 Nadia S. Larse Versio of 13 October 2014. 1. Costructio of the product measure The purpose of these otes is to preset the

More information

Integrable Functions. { f n } is called a determining sequence for f. If f is integrable with respect to, then f d does exist as a finite real number

Integrable Functions. { f n } is called a determining sequence for f. If f is integrable with respect to, then f d does exist as a finite real number MATH 532 Itegrable Fuctios Dr. Neal, WKU We ow shall defie what it meas for a measurable fuctio to be itegrable, show that all itegral properties of simple fuctios still hold, ad the give some coditios

More information

1 Convergence in Probability and the Weak Law of Large Numbers

1 Convergence in Probability and the Weak Law of Large Numbers 36-752 Advaced Probability Overview Sprig 2018 8. Covergece Cocepts: i Probability, i L p ad Almost Surely Istructor: Alessadro Rialdo Associated readig: Sec 2.4, 2.5, ad 4.11 of Ash ad Doléas-Dade; Sec

More information

Chapter 3. Strong convergence. 3.1 Definition of almost sure convergence

Chapter 3. Strong convergence. 3.1 Definition of almost sure convergence Chapter 3 Strog covergece As poited out i the Chapter 2, there are multiple ways to defie the otio of covergece of a sequece of radom variables. That chapter defied covergece i probability, covergece i

More information

Chapter 6 Infinite Series

Chapter 6 Infinite Series Chapter 6 Ifiite Series I the previous chapter we cosidered itegrals which were improper i the sese that the iterval of itegratio was ubouded. I this chapter we are goig to discuss a topic which is somewhat

More information

Math 525: Lecture 5. January 18, 2018

Math 525: Lecture 5. January 18, 2018 Math 525: Lecture 5 Jauary 18, 2018 1 Series (review) Defiitio 1.1. A sequece (a ) R coverges to a poit L R (writte a L or lim a = L) if for each ǫ > 0, we ca fid N such that a L < ǫ for all N. If the

More information

7.1 Convergence of sequences of random variables

7.1 Convergence of sequences of random variables Chapter 7 Limit Theorems Throughout this sectio we will assume a probability space (, F, P), i which is defied a ifiite sequece of radom variables (X ) ad a radom variable X. The fact that for every ifiite

More information

7.1 Convergence of sequences of random variables

7.1 Convergence of sequences of random variables Chapter 7 Limit theorems Throughout this sectio we will assume a probability space (Ω, F, P), i which is defied a ifiite sequece of radom variables (X ) ad a radom variable X. The fact that for every ifiite

More information

Sequences and Series of Functions

Sequences and Series of Functions Chapter 6 Sequeces ad Series of Fuctios 6.1. Covergece of a Sequece of Fuctios Poitwise Covergece. Defiitio 6.1. Let, for each N, fuctio f : A R be defied. If, for each x A, the sequece (f (x)) coverges

More information

Solution. 1 Solutions of Homework 1. Sangchul Lee. October 27, Problem 1.1

Solution. 1 Solutions of Homework 1. Sangchul Lee. October 27, Problem 1.1 Solutio Sagchul Lee October 7, 017 1 Solutios of Homework 1 Problem 1.1 Let Ω,F,P) be a probability space. Show that if {A : N} F such that A := lim A exists, the PA) = lim PA ). Proof. Usig the cotiuity

More information

Lecture 19: Convergence

Lecture 19: Convergence Lecture 19: Covergece Asymptotic approach I statistical aalysis or iferece, a key to the success of fidig a good procedure is beig able to fid some momets ad/or distributios of various statistics. I may

More information

MATH 413 FINAL EXAM. f(x) f(y) M x y. x + 1 n

MATH 413 FINAL EXAM. f(x) f(y) M x y. x + 1 n MATH 43 FINAL EXAM Math 43 fial exam, 3 May 28. The exam starts at 9: am ad you have 5 miutes. No textbooks or calculators may be used durig the exam. This exam is prited o both sides of the paper. Good

More information

Singular Continuous Measures by Michael Pejic 5/14/10

Singular Continuous Measures by Michael Pejic 5/14/10 Sigular Cotiuous Measures by Michael Peic 5/4/0 Prelimiaries Give a set X, a σ-algebra o X is a collectio of subsets of X that cotais X ad ad is closed uder complemetatio ad coutable uios hece, coutable

More information

Advanced Stochastic Processes.

Advanced Stochastic Processes. Advaced Stochastic Processes. David Gamarik LECTURE 2 Radom variables ad measurable fuctios. Strog Law of Large Numbers (SLLN). Scary stuff cotiued... Outlie of Lecture Radom variables ad measurable fuctios.

More information

If a subset E of R contains no open interval, is it of zero measure? For instance, is the set of irrationals in [0, 1] is of measure zero?

If a subset E of R contains no open interval, is it of zero measure? For instance, is the set of irrationals in [0, 1] is of measure zero? 2 Lebesgue Measure I Chapter 1 we defied the cocept of a set of measure zero, ad we have observed that every coutable set is of measure zero. Here are some atural questios: If a subset E of R cotais a

More information

The Boolean Ring of Intervals

The Boolean Ring of Intervals MATH 532 Lebesgue Measure Dr. Neal, WKU We ow shall apply the results obtaied about outer measure to the legth measure o the real lie. Throughout, our space X will be the set of real umbers R. Whe ecessary,

More information

Introduction to Probability. Ariel Yadin

Introduction to Probability. Ariel Yadin Itroductio to robability Ariel Yadi Lecture 2 *** Ja. 7 ***. Covergece of Radom Variables As i the case of sequeces of umbers, we would like to talk about covergece of radom variables. There are may ways

More information

MASSACHUSETTS INSTITUTE OF TECHNOLOGY 6.436J/15.085J Fall 2008 Lecture 19 11/17/2008 LAWS OF LARGE NUMBERS II THE STRONG LAW OF LARGE NUMBERS

MASSACHUSETTS INSTITUTE OF TECHNOLOGY 6.436J/15.085J Fall 2008 Lecture 19 11/17/2008 LAWS OF LARGE NUMBERS II THE STRONG LAW OF LARGE NUMBERS MASSACHUSTTS INSTITUT OF TCHNOLOGY 6.436J/5.085J Fall 2008 Lecture 9 /7/2008 LAWS OF LARG NUMBRS II Cotets. The strog law of large umbers 2. The Cheroff boud TH STRONG LAW OF LARG NUMBRS While the weak

More information

MASSACHUSETTS INSTITUTE OF TECHNOLOGY 6.265/15.070J Fall 2013 Lecture 21 11/27/2013

MASSACHUSETTS INSTITUTE OF TECHNOLOGY 6.265/15.070J Fall 2013 Lecture 21 11/27/2013 MASSACHUSETTS INSTITUTE OF TECHNOLOGY 6.265/15.070J Fall 2013 Lecture 21 11/27/2013 Fuctioal Law of Large Numbers. Costructio of the Wieer Measure Cotet. 1. Additioal techical results o weak covergece

More information

2 Banach spaces and Hilbert spaces

2 Banach spaces and Hilbert spaces 2 Baach spaces ad Hilbert spaces Tryig to do aalysis i the ratioal umbers is difficult for example cosider the set {x Q : x 2 2}. This set is o-empty ad bouded above but does ot have a least upper boud

More information

M17 MAT25-21 HOMEWORK 5 SOLUTIONS

M17 MAT25-21 HOMEWORK 5 SOLUTIONS M17 MAT5-1 HOMEWORK 5 SOLUTIONS 1. To Had I Cauchy Codesatio Test. Exercise 1: Applicatio of the Cauchy Codesatio Test Use the Cauchy Codesatio Test to prove that 1 diverges. Solutio 1. Give the series

More information

MAS111 Convergence and Continuity

MAS111 Convergence and Continuity MAS Covergece ad Cotiuity Key Objectives At the ed of the course, studets should kow the followig topics ad be able to apply the basic priciples ad theorems therei to solvig various problems cocerig covergece

More information

A Proof of Birkhoff s Ergodic Theorem

A Proof of Birkhoff s Ergodic Theorem A Proof of Birkhoff s Ergodic Theorem Joseph Hora September 2, 205 Itroductio I Fall 203, I was learig the basics of ergodic theory, ad I came across this theorem. Oe of my supervisors, Athoy Quas, showed

More information

Probability for mathematicians INDEPENDENCE TAU

Probability for mathematicians INDEPENDENCE TAU Probability for mathematicias INDEPENDENCE TAU 2013 28 Cotets 3 Ifiite idepedet sequeces 28 3a Idepedet evets........................ 28 3b Idepedet radom variables.................. 33 3 Ifiite idepedet

More information

Introduction to Probability. Ariel Yadin. Lecture 2

Introduction to Probability. Ariel Yadin. Lecture 2 Itroductio to Probability Ariel Yadi Lecture 2 1. Discrete Probability Spaces Discrete probability spaces are those for which the sample space is coutable. We have already see that i this case we ca take

More information

Measure and Measurable Functions

Measure and Measurable Functions 3 Measure ad Measurable Fuctios 3.1 Measure o a Arbitrary σ-algebra Recall from Chapter 2 that the set M of all Lebesgue measurable sets has the followig properties: R M, E M implies E c M, E M for N implies

More information

Definition 4.2. (a) A sequence {x n } in a Banach space X is a basis for X if. unique scalars a n (x) such that x = n. a n (x) x n. (4.

Definition 4.2. (a) A sequence {x n } in a Banach space X is a basis for X if. unique scalars a n (x) such that x = n. a n (x) x n. (4. 4. BASES I BAACH SPACES 39 4. BASES I BAACH SPACES Sice a Baach space X is a vector space, it must possess a Hamel, or vector space, basis, i.e., a subset {x γ } γ Γ whose fiite liear spa is all of X ad

More information

An Introduction to Randomized Algorithms

An Introduction to Randomized Algorithms A Itroductio to Radomized Algorithms The focus of this lecture is to study a radomized algorithm for quick sort, aalyze it usig probabilistic recurrece relatios, ad also provide more geeral tools for aalysis

More information

Probability and Random Processes

Probability and Random Processes Probability ad Radom Processes Lecture 5 Probability ad radom variables The law of large umbers Mikael Skoglud, Probability ad radom processes 1/21 Why Measure Theoretic Probability? Stroger limit theorems

More information

Solutions to HW Assignment 1

Solutions to HW Assignment 1 Solutios to HW: 1 Course: Theory of Probability II Page: 1 of 6 Uiversity of Texas at Austi Solutios to HW Assigmet 1 Problem 1.1. Let Ω, F, {F } 0, P) be a filtered probability space ad T a stoppig time.

More information

Chapter IV Integration Theory

Chapter IV Integration Theory Chapter IV Itegratio Theory Lectures 32-33 1. Costructio of the itegral I this sectio we costruct the abstract itegral. As a matter of termiology, we defie a measure space as beig a triple (, A, µ), where

More information

ST5215: Advanced Statistical Theory

ST5215: Advanced Statistical Theory ST525: Advaced Statistical Theory Departmet of Statistics & Applied Probability Tuesday, September 7, 2 ST525: Advaced Statistical Theory Lecture : The law of large umbers The Law of Large Numbers The

More information

(A sequence also can be thought of as the list of function values attained for a function f :ℵ X, where f (n) = x n for n 1.) x 1 x N +k x N +4 x 3

(A sequence also can be thought of as the list of function values attained for a function f :ℵ X, where f (n) = x n for n 1.) x 1 x N +k x N +4 x 3 MATH 337 Sequeces Dr. Neal, WKU Let X be a metric space with distace fuctio d. We shall defie the geeral cocept of sequece ad limit i a metric space, the apply the results i particular to some special

More information

REAL ANALYSIS II: PROBLEM SET 1 - SOLUTIONS

REAL ANALYSIS II: PROBLEM SET 1 - SOLUTIONS REAL ANALYSIS II: PROBLEM SET 1 - SOLUTIONS 18th Feb, 016 Defiitio (Lipschitz fuctio). A fuctio f : R R is said to be Lipschitz if there exists a positive real umber c such that for ay x, y i the domai

More information

Lecture 8: Convergence of transformations and law of large numbers

Lecture 8: Convergence of transformations and law of large numbers Lecture 8: Covergece of trasformatios ad law of large umbers Trasformatio ad covergece Trasformatio is a importat tool i statistics. If X coverges to X i some sese, we ofte eed to check whether g(x ) coverges

More information

Math Solutions to homework 6

Math Solutions to homework 6 Math 175 - Solutios to homework 6 Cédric De Groote November 16, 2017 Problem 1 (8.11 i the book): Let K be a compact Hermitia operator o a Hilbert space H ad let the kerel of K be {0}. Show that there

More information

7 Sequences of real numbers

7 Sequences of real numbers 40 7 Sequeces of real umbers 7. Defiitios ad examples Defiitio 7... A sequece of real umbers is a real fuctio whose domai is the set N of atural umbers. Let s : N R be a sequece. The the values of s are

More information

lim za n n = z lim a n n.

lim za n n = z lim a n n. Lecture 6 Sequeces ad Series Defiitio 1 By a sequece i a set A, we mea a mappig f : N A. It is customary to deote a sequece f by {s } where, s := f(). A sequece {z } of (complex) umbers is said to be coverget

More information

Introduction to Probability. Ariel Yadin. Lecture 7

Introduction to Probability. Ariel Yadin. Lecture 7 Itroductio to Probability Ariel Yadi Lecture 7 1. Idepedece Revisited 1.1. Some remiders. Let (Ω, F, P) be a probability space. Give a collectio of subsets K F, recall that the σ-algebra geerated by K,

More information

Axioms of Measure Theory

Axioms of Measure Theory MATH 532 Axioms of Measure Theory Dr. Neal, WKU I. The Space Throughout the course, we shall let X deote a geeric o-empty set. I geeral, we shall ot assume that ay algebraic structure exists o X so that

More information

Lecture Notes for Analysis Class

Lecture Notes for Analysis Class Lecture Notes for Aalysis Class Topological Spaces A topology for a set X is a collectio T of subsets of X such that: (a) X ad the empty set are i T (b) Uios of elemets of T are i T (c) Fiite itersectios

More information

s = and t = with C ij = A i B j F. (i) Note that cs = M and so ca i µ(a i ) I E (cs) = = c a i µ(a i ) = ci E (s). (ii) Note that s + t = M and so

s = and t = with C ij = A i B j F. (i) Note that cs = M and so ca i µ(a i ) I E (cs) = = c a i µ(a i ) = ci E (s). (ii) Note that s + t = M and so 3 From the otes we see that the parts of Theorem 4. that cocer us are: Let s ad t be two simple o-egative F-measurable fuctios o X, F, µ ad E, F F. The i I E cs ci E s for all c R, ii I E s + t I E s +

More information

1+x 1 + α+x. x = 2(α x2 ) 1+x

1+x 1 + α+x. x = 2(α x2 ) 1+x Math 2030 Homework 6 Solutios # [Problem 5] For coveiece we let α lim sup a ad β lim sup b. Without loss of geerality let us assume that α β. If α the by assumptio β < so i this case α + β. By Theorem

More information

It is always the case that unions, intersections, complements, and set differences are preserved by the inverse image of a function.

It is always the case that unions, intersections, complements, and set differences are preserved by the inverse image of a function. MATH 532 Measurable Fuctios Dr. Neal, WKU Throughout, let ( X, F, µ) be a measure space ad let (!, F, P ) deote the special case of a probability space. We shall ow begi to study real-valued fuctios defied

More information

BIRKHOFF ERGODIC THEOREM

BIRKHOFF ERGODIC THEOREM BIRKHOFF ERGODIC THEOREM Abstract. We will give a proof of the poitwise ergodic theorem, which was first proved by Birkhoff. May improvemets have bee made sice Birkhoff s orgial proof. The versio we give

More information

STAT Homework 1 - Solutions

STAT Homework 1 - Solutions STAT-36700 Homework 1 - Solutios Fall 018 September 11, 018 This cotais solutios for Homework 1. Please ote that we have icluded several additioal commets ad approaches to the problems to give you better

More information

6.3 Testing Series With Positive Terms

6.3 Testing Series With Positive Terms 6.3. TESTING SERIES WITH POSITIVE TERMS 307 6.3 Testig Series With Positive Terms 6.3. Review of what is kow up to ow I theory, testig a series a i for covergece amouts to fidig the i= sequece of partial

More information

Infinite Sequences and Series

Infinite Sequences and Series Chapter 6 Ifiite Sequeces ad Series 6.1 Ifiite Sequeces 6.1.1 Elemetary Cocepts Simply speakig, a sequece is a ordered list of umbers writte: {a 1, a 2, a 3,...a, a +1,...} where the elemets a i represet

More information

1 Lecture 2: Sequence, Series and power series (8/14/2012)

1 Lecture 2: Sequence, Series and power series (8/14/2012) Summer Jump-Start Program for Aalysis, 202 Sog-Yig Li Lecture 2: Sequece, Series ad power series (8/4/202). More o sequeces Example.. Let {x } ad {y } be two bouded sequeces. Show lim sup (x + y ) lim

More information

4. Partial Sums and the Central Limit Theorem

4. Partial Sums and the Central Limit Theorem 1 of 10 7/16/2009 6:05 AM Virtual Laboratories > 6. Radom Samples > 1 2 3 4 5 6 7 4. Partial Sums ad the Cetral Limit Theorem The cetral limit theorem ad the law of large umbers are the two fudametal theorems

More information

FUNDAMENTALS OF REAL ANALYSIS by. V.1. Product measures

FUNDAMENTALS OF REAL ANALYSIS by. V.1. Product measures FUNDAMENTALS OF REAL ANALSIS by Doğa Çömez V. PRODUCT MEASURE SPACES V.1. Product measures Let (, A, µ) ad (, B, ν) be two measure spaces. I this sectio we will costruct a product measure µ ν o that coicides

More information

f n (x) f m (x) < ɛ/3 for all x A. By continuity of f n and f m we can find δ > 0 such that d(x, x 0 ) < δ implies that

f n (x) f m (x) < ɛ/3 for all x A. By continuity of f n and f m we can find δ > 0 such that d(x, x 0 ) < δ implies that Lecture 15 We have see that a sequece of cotiuous fuctios which is uiformly coverget produces a limit fuctio which is also cotiuous. We shall stregthe this result ow. Theorem 1 Let f : X R or (C) be a

More information

EE 4TM4: Digital Communications II Probability Theory

EE 4TM4: Digital Communications II Probability Theory 1 EE 4TM4: Digital Commuicatios II Probability Theory I. RANDOM VARIABLES A radom variable is a real-valued fuctio defied o the sample space. Example: Suppose that our experimet cosists of tossig two fair

More information

Analytic Continuation

Analytic Continuation Aalytic Cotiuatio The stadard example of this is give by Example Let h (z) = 1 + z + z 2 + z 3 +... kow to coverge oly for z < 1. I fact h (z) = 1/ (1 z) for such z. Yet H (z) = 1/ (1 z) is defied for

More information

University of Colorado Denver Dept. Math. & Stat. Sciences Applied Analysis Preliminary Exam 13 January 2012, 10:00 am 2:00 pm. Good luck!

University of Colorado Denver Dept. Math. & Stat. Sciences Applied Analysis Preliminary Exam 13 January 2012, 10:00 am 2:00 pm. Good luck! Uiversity of Colorado Dever Dept. Math. & Stat. Scieces Applied Aalysis Prelimiary Exam 13 Jauary 01, 10:00 am :00 pm Name: The proctor will let you read the followig coditios before the exam begis, ad

More information

Distribution of Random Samples & Limit theorems

Distribution of Random Samples & Limit theorems STAT/MATH 395 A - PROBABILITY II UW Witer Quarter 2017 Néhémy Lim Distributio of Radom Samples & Limit theorems 1 Distributio of i.i.d. Samples Motivatig example. Assume that the goal of a study is to

More information

Chapter 0. Review of set theory. 0.1 Sets

Chapter 0. Review of set theory. 0.1 Sets Chapter 0 Review of set theory Set theory plays a cetral role i the theory of probability. Thus, we will ope this course with a quick review of those otios of set theory which will be used repeatedly.

More information

Fall 2013 MTH431/531 Real analysis Section Notes

Fall 2013 MTH431/531 Real analysis Section Notes Fall 013 MTH431/531 Real aalysis Sectio 8.1-8. Notes Yi Su 013.11.1 1. Defiitio of uiform covergece. We look at a sequece of fuctios f (x) ad study the coverget property. Notice we have two parameters

More information

1 Introduction. 1.1 Notation and Terminology

1 Introduction. 1.1 Notation and Terminology 1 Itroductio You have already leared some cocepts of calculus such as limit of a sequece, limit, cotiuity, derivative, ad itegral of a fuctio etc. Real Aalysis studies them more rigorously usig a laguage

More information

Theorem 3. A subset S of a topological space X is compact if and only if every open cover of S by open sets in X has a finite subcover.

Theorem 3. A subset S of a topological space X is compact if and only if every open cover of S by open sets in X has a finite subcover. Compactess Defiitio 1. A cover or a coverig of a topological space X is a family C of subsets of X whose uio is X. A subcover of a cover C is a subfamily of C which is a cover of X. A ope cover of X is

More information

ECE 330:541, Stochastic Signals and Systems Lecture Notes on Limit Theorems from Probability Fall 2002

ECE 330:541, Stochastic Signals and Systems Lecture Notes on Limit Theorems from Probability Fall 2002 ECE 330:541, Stochastic Sigals ad Systems Lecture Notes o Limit Theorems from robability Fall 00 I practice, there are two ways we ca costruct a ew sequece of radom variables from a old sequece of radom

More information

HOMEWORK #4 - MA 504

HOMEWORK #4 - MA 504 HOMEWORK #4 - MA 504 PAULINHO TCHATCHATCHA Chapter 2, problem 19. (a) If A ad B are disjoit closed sets i some metric space X, prove that they are separated. (b) Prove the same for disjoit ope set. (c)

More information

FUNDAMENTALS OF REAL ANALYSIS by

FUNDAMENTALS OF REAL ANALYSIS by FUNDAMENTALS OF REAL ANALYSIS by Doğa Çömez Backgroud: All of Math 450/1 material. Namely: basic set theory, relatios ad PMI, structure of N, Z, Q ad R, basic properties of (cotiuous ad differetiable)

More information

MAT1026 Calculus II Basic Convergence Tests for Series

MAT1026 Calculus II Basic Convergence Tests for Series MAT026 Calculus II Basic Covergece Tests for Series Egi MERMUT 202.03.08 Dokuz Eylül Uiversity Faculty of Sciece Departmet of Mathematics İzmir/TURKEY Cotets Mootoe Covergece Theorem 2 2 Series of Real

More information

Solutions to home assignments (sketches)

Solutions to home assignments (sketches) Matematiska Istitutioe Peter Kumli 26th May 2004 TMA401 Fuctioal Aalysis MAN670 Applied Fuctioal Aalysis 4th quarter 2003/2004 All documet cocerig the course ca be foud o the course home page: http://www.math.chalmers.se/math/grudutb/cth/tma401/

More information

Lecture 10 October Minimaxity and least favorable prior sequences

Lecture 10 October Minimaxity and least favorable prior sequences STATS 300A: Theory of Statistics Fall 205 Lecture 0 October 22 Lecturer: Lester Mackey Scribe: Brya He, Rahul Makhijai Warig: These otes may cotai factual ad/or typographic errors. 0. Miimaxity ad least

More information

Beurling Integers: Part 2

Beurling Integers: Part 2 Beurlig Itegers: Part 2 Isomorphisms Devi Platt July 11, 2015 1 Prime Factorizatio Sequeces I the last article we itroduced the Beurlig geeralized itegers, which ca be represeted as a sequece of real umbers

More information

n=1 a n is the sequence (s n ) n 1 n=1 a n converges to s. We write a n = s, n=1 n=1 a n

n=1 a n is the sequence (s n ) n 1 n=1 a n converges to s. We write a n = s, n=1 n=1 a n Series. Defiitios ad first properties A series is a ifiite sum a + a + a +..., deoted i short by a. The sequece of partial sums of the series a is the sequece s ) defied by s = a k = a +... + a,. k= Defiitio

More information

Introduction to Functional Analysis

Introduction to Functional Analysis MIT OpeCourseWare http://ocw.mit.edu 18.10 Itroductio to Fuctioal Aalysis Sprig 009 For iformatio about citig these materials or our Terms of Use, visit: http://ocw.mit.edu/terms. LECTURE OTES FOR 18.10,

More information

MATH301 Real Analysis (2008 Fall) Tutorial Note #7. k=1 f k (x) converges pointwise to S(x) on E if and

MATH301 Real Analysis (2008 Fall) Tutorial Note #7. k=1 f k (x) converges pointwise to S(x) on E if and MATH01 Real Aalysis (2008 Fall) Tutorial Note #7 Sequece ad Series of fuctio 1: Poitwise Covergece ad Uiform Covergece Part I: Poitwise Covergece Defiitio of poitwise covergece: A sequece of fuctios f

More information

5 Birkhoff s Ergodic Theorem

5 Birkhoff s Ergodic Theorem 5 Birkhoff s Ergodic Theorem Amog the most useful of the various geeralizatios of KolmogorovâĂŹs strog law of large umbers are the ergodic theorems of Birkhoff ad Kigma, which exted the validity of the

More information

The Borel hierarchy classifies subsets of the reals by their topological complexity. Another approach is to classify them by size.

The Borel hierarchy classifies subsets of the reals by their topological complexity. Another approach is to classify them by size. Lecture 7: Measure ad Category The Borel hierarchy classifies subsets of the reals by their topological complexity. Aother approach is to classify them by size. Filters ad Ideals The most commo measure

More information

Here are some examples of algebras: F α = A(G). Also, if A, B A(G) then A, B F α. F α = A(G). In other words, A(G)

Here are some examples of algebras: F α = A(G). Also, if A, B A(G) then A, B F α. F α = A(G). In other words, A(G) MATH 529 Probability Axioms Here we shall use the geeral axioms of a probability measure to derive several importat results ivolvig probabilities of uios ad itersectios. Some more advaced results will

More information

Archimedes - numbers for counting, otherwise lengths, areas, etc. Kepler - geometry for planetary motion

Archimedes - numbers for counting, otherwise lengths, areas, etc. Kepler - geometry for planetary motion Topics i Aalysis 3460:589 Summer 007 Itroductio Ree descartes - aalysis (breaig dow) ad sythesis Sciece as models of ature : explaatory, parsimoious, predictive Most predictios require umerical values,

More information

Notes #3 Sequences Limit Theorems Monotone and Subsequences Bolzano-WeierstraßTheorem Limsup & Liminf of Sequences Cauchy Sequences and Completeness

Notes #3 Sequences Limit Theorems Monotone and Subsequences Bolzano-WeierstraßTheorem Limsup & Liminf of Sequences Cauchy Sequences and Completeness Notes #3 Sequeces Limit Theorems Mootoe ad Subsequeces Bolzao-WeierstraßTheorem Limsup & Limif of Sequeces Cauchy Sequeces ad Completeess This sectio of otes focuses o some of the basics of sequeces of

More information

The Wasserstein distances

The Wasserstein distances The Wasserstei distaces March 20, 2011 This documet presets the proof of the mai results we proved o Wasserstei distaces themselves (ad ot o curves i the Wasserstei space). I particular, triagle iequality

More information

Sequences. A Sequence is a list of numbers written in order.

Sequences. A Sequence is a list of numbers written in order. Sequeces A Sequece is a list of umbers writte i order. {a, a 2, a 3,... } The sequece may be ifiite. The th term of the sequece is the th umber o the list. O the list above a = st term, a 2 = 2 d term,

More information

Advanced Analysis. Min Yan Department of Mathematics Hong Kong University of Science and Technology

Advanced Analysis. Min Yan Department of Mathematics Hong Kong University of Science and Technology Advaced Aalysis Mi Ya Departmet of Mathematics Hog Kog Uiversity of Sciece ad Techology September 3, 009 Cotets Limit ad Cotiuity 7 Limit of Sequece 8 Defiitio 8 Property 3 3 Ifiity ad Ifiitesimal 8 4

More information

Sequences and Series

Sequences and Series Sequeces ad Series Sequeces of real umbers. Real umber system We are familiar with atural umbers ad to some extet the ratioal umbers. While fidig roots of algebraic equatios we see that ratioal umbers

More information

Empirical Processes: Glivenko Cantelli Theorems

Empirical Processes: Glivenko Cantelli Theorems Empirical Processes: Gliveko Catelli Theorems Mouliath Baerjee Jue 6, 200 Gliveko Catelli classes of fuctios The reader is referred to Chapter.6 of Weller s Torgo otes, Chapter??? of VDVW ad Chapter 8.3

More information

MASSACHUSETTS INSTITUTE OF TECHNOLOGY 6.265/15.070J Fall 2013 Lecture 3 9/11/2013. Large deviations Theory. Cramér s Theorem

MASSACHUSETTS INSTITUTE OF TECHNOLOGY 6.265/15.070J Fall 2013 Lecture 3 9/11/2013. Large deviations Theory. Cramér s Theorem MASSACHUSETTS INSTITUTE OF TECHNOLOGY 6.265/5.070J Fall 203 Lecture 3 9//203 Large deviatios Theory. Cramér s Theorem Cotet.. Cramér s Theorem. 2. Rate fuctio ad properties. 3. Chage of measure techique.

More information

This section is optional.

This section is optional. 4 Momet Geeratig Fuctios* This sectio is optioal. The momet geeratig fuctio g : R R of a radom variable X is defied as g(t) = E[e tx ]. Propositio 1. We have g () (0) = E[X ] for = 1, 2,... Proof. Therefore

More information

Math 341 Lecture #31 6.5: Power Series

Math 341 Lecture #31 6.5: Power Series Math 341 Lecture #31 6.5: Power Series We ow tur our attetio to a particular kid of series of fuctios, amely, power series, f(x = a x = a 0 + a 1 x + a 2 x 2 + where a R for all N. I terms of a series

More information

Math 299 Supplement: Real Analysis Nov 2013

Math 299 Supplement: Real Analysis Nov 2013 Math 299 Supplemet: Real Aalysis Nov 203 Algebra Axioms. I Real Aalysis, we work withi the axiomatic system of real umbers: the set R alog with the additio ad multiplicatio operatios +,, ad the iequality

More information

Review Problems 1. ICME and MS&E Refresher Course September 19, 2011 B = C = AB = A = A 2 = A 3... C 2 = C 3 = =

Review Problems 1. ICME and MS&E Refresher Course September 19, 2011 B = C = AB = A = A 2 = A 3... C 2 = C 3 = = Review Problems ICME ad MS&E Refresher Course September 9, 0 Warm-up problems. For the followig matrices A = 0 B = C = AB = 0 fid all powers A,A 3,(which is A times A),... ad B,B 3,... ad C,C 3,... Solutio:

More information

Math 61CM - Solutions to homework 3

Math 61CM - Solutions to homework 3 Math 6CM - Solutios to homework 3 Cédric De Groote October 2 th, 208 Problem : Let F be a field, m 0 a fixed oegative iteger ad let V = {a 0 + a x + + a m x m a 0,, a m F} be the vector space cosistig

More information

MATH4822E FOURIER ANALYSIS AND ITS APPLICATIONS

MATH4822E FOURIER ANALYSIS AND ITS APPLICATIONS MATH48E FOURIER ANALYSIS AND ITS APPLICATIONS 7.. Cesàro summability. 7. Summability methods Arithmetic meas. The followig idea is due to the Italia geometer Eresto Cesàro (859-96). He shows that eve if

More information

MASSACHUSETTS INSTITUTE OF TECHNOLOGY 6.265/15.070J Fall 2013 Lecture 2 9/9/2013. Large Deviations for i.i.d. Random Variables

MASSACHUSETTS INSTITUTE OF TECHNOLOGY 6.265/15.070J Fall 2013 Lecture 2 9/9/2013. Large Deviations for i.i.d. Random Variables MASSACHUSETTS INSTITUTE OF TECHNOLOGY 6.265/15.070J Fall 2013 Lecture 2 9/9/2013 Large Deviatios for i.i.d. Radom Variables Cotet. Cheroff boud usig expoetial momet geeratig fuctios. Properties of a momet

More information

Series III. Chapter Alternating Series

Series III. Chapter Alternating Series Chapter 9 Series III With the exceptio of the Null Sequece Test, all the tests for series covergece ad divergece that we have cosidered so far have dealt oly with series of oegative terms. Series with

More information

We are mainly going to be concerned with power series in x, such as. (x)} converges - that is, lims N n

We are mainly going to be concerned with power series in x, such as. (x)} converges - that is, lims N n Review of Power Series, Power Series Solutios A power series i x - a is a ifiite series of the form c (x a) =c +c (x a)+(x a) +... We also call this a power series cetered at a. Ex. (x+) is cetered at

More information

1 = δ2 (0, ), Y Y n nδ. , T n = Y Y n n. ( U n,k + X ) ( f U n,k + Y ) n 2n f U n,k + θ Y ) 2 E X1 2 X1

1 = δ2 (0, ), Y Y n nδ. , T n = Y Y n n. ( U n,k + X ) ( f U n,k + Y ) n 2n f U n,k + θ Y ) 2 E X1 2 X1 8. The cetral limit theorems 8.1. The cetral limit theorem for i.i.d. sequeces. ecall that C ( is N -separatig. Theorem 8.1. Let X 1, X,... be i.i.d. radom variables with EX 1 = ad EX 1 = σ (,. Suppose

More information

Probability 2 - Notes 10. Lemma. If X is a random variable and g(x) 0 for all x in the support of f X, then P(g(X) 1) E[g(X)].

Probability 2 - Notes 10. Lemma. If X is a random variable and g(x) 0 for all x in the support of f X, then P(g(X) 1) E[g(X)]. Probability 2 - Notes 0 Some Useful Iequalities. Lemma. If X is a radom variable ad g(x 0 for all x i the support of f X, the P(g(X E[g(X]. Proof. (cotiuous case P(g(X Corollaries x:g(x f X (xdx x:g(x

More information

Assignment 5: Solutions

Assignment 5: Solutions McGill Uiversity Departmet of Mathematics ad Statistics MATH 54 Aalysis, Fall 05 Assigmet 5: Solutios. Let y be a ubouded sequece of positive umbers satisfyig y + > y for all N. Let x be aother sequece

More information

Lecture 10: Bounded Linear Operators and Orthogonality in Hilbert Spaces

Lecture 10: Bounded Linear Operators and Orthogonality in Hilbert Spaces Lecture : Bouded Liear Operators ad Orthogoality i Hilbert Spaces 34 Bouded Liear Operator Let ( X, ), ( Y, ) i i be ored liear vector spaces ad { } X Y The, T is said to be bouded if a real uber c such

More information

Math 113 Exam 3 Practice

Math 113 Exam 3 Practice Math Exam Practice Exam will cover.-.9. This sheet has three sectios. The first sectio will remid you about techiques ad formulas that you should kow. The secod gives a umber of practice questios for you

More information

On Random Line Segments in the Unit Square

On Random Line Segments in the Unit Square O Radom Lie Segmets i the Uit Square Thomas A. Courtade Departmet of Electrical Egieerig Uiversity of Califoria Los Ageles, Califoria 90095 Email: tacourta@ee.ucla.edu I. INTRODUCTION Let Q = [0, 1] [0,

More information

Detailed proofs of Propositions 3.1 and 3.2

Detailed proofs of Propositions 3.1 and 3.2 Detailed proofs of Propositios 3. ad 3. Proof of Propositio 3. NB: itegratio sets are geerally omitted for itegrals defied over a uit hypercube [0, s with ay s d. We first give four lemmas. The proof of

More information