A CLASS OF SELF-AFFINE SETS AND SELF-AFFINE MEASURES. 1. Introduction

Size: px
Start display at page:

Download "A CLASS OF SELF-AFFINE SETS AND SELF-AFFINE MEASURES. 1. Introduction"

Transcription

1 A CLASS OF SELF-AFFINE SETS AND SELF-AFFINE MEASURES DE-JUN FENG AND YANG WANG 1 Itroductio Let I = {φ m be a iterated fuctio system (IFS cosistig of a family of cotractive affie maps o R d Hutchiso [13] proved that there exists a uique compact set K = K(I, called the attractor of the IFS I, such that K = m φ (K Moreover, for ay give probability vector p = (p 1,, p m, ie p > 0 for all ad m p = 1, there exists a uique compactly supported probability measure ν = ν I,p such that (11 ν = p ν φ 1 This paper is devoted to the study of fudametal properties of a class of self-affie sets ad measures, such as the L q spectrum, the Hausdorff dimesio ad the etropy dimesio It is well kow that problems cocerig self-affie sets ad measures are typically difficult Questios that may be trivial i the self-similar settig are ofte itractable i the self-affie settig A tellig example is calculatig the Hausdorff ad box dimesios of the attractor of a IFS I = {φ m If all φ are similitudes ad I satisfies the so-called ope set coditio (OSC the Haudorff dimesio ad the box dimesio of the attractor K(I agree, ad they are easily computable by the formula ρ dim H(K = 1 where ρ i deotes the cotractio ratio of φ, see eg Falcoer [5] Eve without the ope set coditio the dimesio of K(I ca ofte be computed if I belogs to a more geeral class called the fiite type IFS, see eg Lalley [16] ad Ngai ad Wag [23] ad the refereces therei However this is o loger the case whe φ i are affie maps Eve uder the ope The first author is supported i part by the Special Fuds for Maor State Basic Research Proects i Chia The secod author is supported i part by the Natioal Sciece Foudatio, grats DMS ad DMS

2 2 DE-JUN FENG AND YANG WANG set coditio we kow how to compute the Hausdorff dimesio of K(I oly for very special I s, ad for which the solutios are quite otrivial McMulle [21] ad Bedford [1] idepedetly computed the Hausdorff ad box dimesios of K(I for I = {φ m i which all φ have the form (12 φ (x = [ ] k 1 x + [ ] a / b /k where all a, b are itegers, 0 a < ad 0 b < m They foud that the Hausdorff dimesio ad the box dimesio are ot the same i geeral Lalley ad Gatzouras [17], i a highly techical paper alog the same spirit of [21], computed the Hausdorff ad box dimesios for a broader class of IFS I = {φ m, i which φ map the uit square (0, 1 2 ito disoit rectagles havig certai geometric arragemet iside the uit square More precisely, i the Lalley-Gatzouras class all rectagles φ ((0, 1 2 are parallel to the axes ad have loger sides parallel to the x-axis Furthermore oce proected oto the x-axis these rectagles are either idetical or disoit Aside from a few other special cases such as the graph-directed McMulle class studied by Keyo ad Peres [14], the Lalley-Gatzouras class (which icludes the McMulle class remais the oly determiistic class of self-affie sets whose Hausdorff dimesio are kow I [10] Hu obtaied the box dimesio of a class of oliear self-affie sets i terms of the topological pressure More precisely, he cosidered expadig maps o R 2 which leaves ivariat a strog ustable foliatio ad obtaied a formula for the box-dimesio of its closed ivariats sets ivolvig topological pressures Alog aother directio, Falcoer, i a celebrated paper [4], gave a variatioal formula for the Hausdorff ad box dimesios for almost all self-affie sets uder some assumptios Later, Hueter ad Lalley [12] ad Solomyak [27] proved that Falcoer s formula remais true uder some weaker coditios We focus o the L q spectrum ad the Hausdorff ad etropy dimesios of a self-affie measure i this paper These quatities are importat basic igrediets i the study of fractal geometry, particularly i the study of multifractal pheomea As a by-product we also obtai results o dimesios of self-affie sets Let ν be a compactly supported measure i R d ad q R For each 1 let D be the set of cubes {[0, 2 d + α : α 2 Z d The L q spectrum of ν is defied as log τ (ν, q (13 τ(ν, q = lim log 2, where τ (ν, q = Q D (ν(q q,

3 A CLASS OF SELF-AFFINE SETS AND SELF-AFFINE MEASURES 3 if the limit exists Related to τ(ν, q are the L q dimesio D(ν, q ad the etropy dimesio h(ν of ν, defied respectively by (14 D(ν, q := ad (15 h(ν := lim τ(ν, q q 1 Q D ν(q log(1/ν(q log 2 if the limit exists For a similarity IFS I = {φ m with the ope set coditio ad ay probability vector p = (p 1,, p m the L q spectrum of ν = ν I,p is kow to be aalytic i q R, give by the equatio (16 p q ρ τ(ν,q = 1, where ρ deotes the cotractio ratio of φ, see Cawley ad Mauldi [2] ad Olse [24] Moreover, the Legedre trasform τ (ν, α of τ(ν, q give by { (17 τ (ν, α := if qα τ(ν, q : q R equals the Hausdorff dimesio of the set { K(α := x supp (ν : ( log ν Br (x lim r 0 + log r = α For a self-similar measure without the ope set coditio, however, the L q spectrum is geerally difficult to obtai ad is calculated for oly a few special cases, see [18, 7, 8, 19] Oe importat such special case is the class of fiite type IFS s ([23], a substatially larger class tha the class with the OSC For a fiite type IFS i R, Feg [9] expressed τ(ν, q via products of certai oegative matrices, ad proved that τ(ν, q is differetiable for q (0, As oe would expect, eve less is kow about the L q spectrum ad the Hausdorff ad etropy dimesios of a self-affie measure Kig [15] calculated τ(ν, q for ν = ν I,p where the IFS I is i the McMulle class (12 He gave a detailed multifractal aalysis for such measures Olse [25] geeralized Kig s results to dimesios d 3 Peres ad Solomyak [26] proved the existece of τ(ν, q ad h(ν for the class of self-coformal measures, ad asked whether they also exist for all self-affie measures I [6], Falcoer gave a variatioal formula for the L q -spectrum (1 < q < 2 for almost all self-affie sets uder some assumptios

4 4 DE-JUN FENG AND YANG WANG I this paper we calculate the L q spectrum ad the etropy dimesio for a class of self-affie measures i R 2 This class of self-affie measures ν I,p requires oly that the uderlyig IFS s I = {φ m satisfy the rectagular ope set coditio (ROSC It is a much larger class tha the McMulle class studied i [15] ad the Lalley-Gatzouras class Simply speakig, I = {φ m i R2 satisfies the ROSC if there is a ope rectagle T such that the maps φ i map T ito disoit rectagles parallel to the axes iside T As a applicatio we obtai the formula for the box dimesio of K(I uder the ROSC as well as the Hausdorff dimesio of ν I,p uder some additioal assumptios Our results o the box dimesio ca be viewed as a extesio of the box dimesio results by Lalley ad Gatzouras [17] ad Hu [10] 2 Statemet of Mai Results We first itroduce some defiitios ad otatios The ambiet dimesio i the rest of the paper will be set to d = 2, although most of the defiitios exted easily to higher dimesios Let I = {φ m be a affie IFS i R2 Throughout this paper we shall always assume that φ (x, y = (a x + c, b y + d with 0 < a, b < 1 for all Thus each φ maps ay rectagle (0, R 1 (0, R 2 + v to a rectagle parallel to the axes Defiitio 21 We say that I = {φ m satisfies the rectagular ope set coditio (ROSC if there exists a ope rectagle T = (0, R 1 (0, R 2 + v such that {φ (T m are disoit subsets of T For a self-affie measure ν = ν I,p associated with I ad probability vector p we shall defie the proectios ν x ad ν y of ν oto the x- ad y-axes, which we rely o heavily i this paper Let I x := {π x φ π 1 x = a x + c ad I y := {π y φ π 1 y = b y + d be the proectios of I, where π x ad π y are the caoical proectios of R 2 oto the x- ad y-axes, respectively We defie ν x = ν I x,p ad ν y = ν I y,p It is easy to check that ν x = ν π 1 x (21 Γ(d := ad ν y = ν πy 1 For ay d = (d 1, d 2,, d m we use Γ(d to deote { t = (t 1, t 2,, t m : t 0, m t = 1, m d t 0 We also use log d to deote (log d 1, log d 2,, log d m (if all d > 0 Our mai theorem cocerig the L q spectrum of ν is:

5 A CLASS OF SELF-AFFINE SETS AND SELF-AFFINE MEASURES 5 Theorem 21 Let I = {φ m be a affie IFS i R2 satisfyig the ROSC, with φ (x, y = (a x + c, b y + d ad 0 < a, b < 1 for all Let a = (a 1, a 2,, a m ad b = (b 1, b 2,, b m Suppose that p = (p 1, p 2,, p m is ay probability vector The for q > 0 the L q spectrum of ν = ν I,p is τ(ν, q = mi(θ a, θ b, where t ( log t τ(ν y, q(log b log a + q log p θ a = if t Γ(log b log a t log a t θ b = if t Γ(log a log b ( log t τ(ν x, q(log a log b + q log p t log b We poit out that if a < b for all the the set Γ(e b is empty, ad if so we have τ(ν, q = θ a I fact we prove: Theorem 22 Uder the hypotheses of Theorem 21, assume furthermore that a b for all The τ(ν, q satisfies (22 a τ(νy,q τ(ν,q b τ(νy,q p q = 1 Theorem 22 allows us to easily calculate the L q spectrum if τ(ν y, q is kow, which is the case if I is i the McMulle or Lalley-Gatzouras class Moreoever, Theorem 21 allows us to calculate τ(ν, q, at least i theory, if the proectios of I oto the two axes are of fiite type by the result of Feg [9], makig the L q spectrum computable for a cosiderably larger class of IFS s tha the Lalley-Gatzouras class Oe of the applicatios of the above two theorems is a formula for the box dimesio of K = K(I It is easy to see that K = supp (ν ad by defiitio dim B (K is simply τ(ν, 0 Therefore we also obtai as a by-product of Theorems 21 ad 22 a formula for dim B (K: Corollary 23 Uder the hypotheses of Theorem 21, we have dim B (K = max(u a, u b, where t (log t dim B (π y (K(log b log a u a = sup t Γ(log b log a t log a ( t log t dim B (π x (K(log a log b u b = sup t Γ(log a log b t log b

6 6 DE-JUN FENG AND YANG WANG If assume furthermore that a b for all, the dim B (K satisfies (23 a dim B(K dim B (π y(k b dim B(π y(k = 1 I particular, if dim B (π y (K = 1 furthermore, the (24 a dim B(K 1 b = 1 The computable cases i the above corollary for dim B (K clearly iclude the Lalley- Gatzouras class, i which a b ad I y satisfies the OSC I [10] Hu obtaied a formula for the box-dimesio dim B (K i terms of topological pressures whe a b uiformly I a persoal commuicatio, Hu iformed us that (23 ca be also derived from his formula after some otrivial calculatios Aother applicatio of the theorems is computig the Hausdorff dimesio of a self-affie measure Let ν be a fiite Borel measure i R d It is said to be exactly dimesioal if there exists a costat c such that log ν(b r (x lim = c ν ae x R d r 0 log r Ngai [22] proved that if τ(ν, q is differetiable at q = 1 the ν is exactly dimesioal, ad dim H (ν = c = d dq τ(ν, 1 As a corollary of Theorem 22 we obtai a Ledrappier-Youg type formula (see [20] for dim H (ν: Theorem 24 Uder the hypotheses of Theorem 22, if τ(ν y, q is differetiable at q = 1 the so is τ(ν, q, ad dim H (ν = p (log p + dim H (ν y (log a log b p log a I particular if furthermore a = a ad b = b for all ad a b the dim H (ν = p log p + dim H(ν y log(a/b log a Our techique ca also be used to study the etropy dimesio, which for a Borel measure ν is defied i (15 It is kow [26] that the etropy dimesio exists for all self-similar (i fact self-coformal measures We determie the etropy dimesio for the self-affie measures with ROSC:

7 A CLASS OF SELF-AFFINE SETS AND SELF-AFFINE MEASURES 7 Theorem 25 Uder the hypotheses of Theorem 21, h(ν = h(ν y p (log b log a + p log p p log a h(ν x p (log a log b + p log p p log b if p (log b log a 0, otherwise 3 Some Combiatorial Results We establish two combiatorial results that will be eeded to prove our mai theorems i this paper First let us itroduce some otatios o symbolic spaces These otatios are mostly stadard We use Σ = Σ(m to deote the alphabet {1, 2,, m Wheever there is o ambiguity we shall use Σ rather tha Σ(m, as m is usually fixed i this paper The set of all words i Σ of legth is deoted by Σ, with Σ := 0 Σ ad Σ N beig the set of all oe-sided ifiite words Here we adopt the covetio that Σ 0 cotais oly the empty word Associated with Σ are two actios: The left shift actio σ ad ad the right shift actio δ, defied respectively by σ( = δ( = ad σ(i 1 i 2 i k = i 2 i k, δ(i 1 i 2 i k = i 1 i k 1 for each i 1 i 2 i k Σ with k 1 We shall use boldface letters i,, l to deote elemets i Σ or Σ N For each sequece a = (a m we may exted it to a fuctio f a : Σ R by f a ( = 1 ad f a (i = a i1 a ik for i = i 1 i k Most of the time, because there is o ambiguity, we shall use the simplified otatio a i i place of f a (i The above are geeral purpose otatios Now we itroduce some that are specific to this paper Suppose that a = (a m ad b = (b m are two sequeces with 0 < a, b < 1 for all For ay 0 < r < 1 let A r := A r (a, b = { i Σ : a δ(i r, b δ(i r, mi (a i, b i < r ad A a r := {i A r : a i b i, A b r := {i A r : a i > b i

8 8 DE-JUN FENG AND YANG WANG Suppose that c = (c m is aother sequece of positive real umbers The obective of this sectio is to evaluate several limits Set ( Θ a = Θ a log i A (c := lim a c i r, ad r 0 + log r ( Θ b = Θ b log i A (c := lim b c i r r 0 + log r Similarly, for ay probability vector p = (p 1, p 2,, p m set Ω a = Ω a i A (c, p := lim a p i log c r i, ad r 0 + log r Ω b = Ω a i A (c, p := lim b p i log c r i r 0 + log r We prove the followig results: Propositio 31 Give sequeces a = (a m ad b = (b m with all a, b i (0, 1 let e a = log b log a ad e b = log a log b (i If a b for some 1 m the Θ a (c exists, ad Θ a t ( log t + log c (c = if, t Γ(log b log a t log a where for ay vector d Γ(d is defied i (21 (ii If a > b for some 1 m the Θ b (c exists, ad Θ b t ( log t + log c (c = if t Γ(log a log b t log b (iii Θ a (c (resp Θ b (c is cotiuous with respect to c if it exists Propositio 32 Uder the assumptios of Propositio 31, ad let p be a probability vector (i If p (log b log a > 0 the (ii If p (log b log a < 0 the (iii If p (log b log a = 0 the lim r 0 + Ω a (c, p = p log c p log a, ad Ωb (c, p = 0 Ω b (c, p = p log c p log b, ad Ωa (c, p = 0 i A a p i log(b r i /a i = lim log r r 0 + i A b p i log(a r i /b i = 0, log r

9 A CLASS OF SELF-AFFINE SETS AND SELF-AFFINE MEASURES 9 ad lim r 0 + i A r p i log c i log r = p log c p log a, We eed to first prove some lemmas For ay i = i 1 i 2 i Σ let [i] Σ N deote the i-cylider [i] := { Σ N : k = i k for 1 k Lemma 33 For ay 0 < r < 1, {[i] : i A r is a partitio of Σ N Proof It is clear that {[i] : i A r are distict subsets i Σ N Furthermore, for ay = Σ N there exists a smallest such that mi(a 1 a, b 1 b < r Therefore [ 1 ] ad [ 1 ] A r This proves the lemma Lemma 34 Let = m with each N The 1 ( log! ( log = t log t + O, 1! 2! m! where t = Proof We apply Stirlig s formula log(q! = q log q q log q + O(1 Thus log(! = log + O(log Sice m is fixed i our settig, m log(! = ( log + O(log = Now log = log(t = log t + log It follows that 1 ( log! = log 1! 2! m! = log + O(log log ( log + O ( log t log t + O For each i = i 1 i 2 i Σ we use i = to deote the legth of i ad i = #{k : i k = to deote the umber of occureces of the letter i i Lemma 35 There exists a costat C > 1 such that C 1 log r 1 i C log r 1 for ay 0 < r < 1 2 ad i A r

10 10 DE-JUN FENG AND YANG WANG Proof Let s + = max {a, b : 1 m ad s = mi {a, b : 1 m The we have s i a i, b i s i + for ay i Σ The lemma follows by settig C = max ( log s, log s log 2 1 Note that the coditio 0 < r < 1 2 ca be replaced with 0 < r < r 0 for ay fixed r 0 < 1 Proof of Propositio 31 We shall prove part (i of the propositio oly, as part (ii follows from a idetical argumet ad part (iii is rather obvious To prove (i we estimate the sum i A a r c i For ay i = i 1 i 2 i A a r we observe that i = i is also i A a, where is ay permutatio of δ(i = i 1 i 1, which gives c i = c i The umber of distict such i is precisely ( 1!/ m! where := δ(i Let T (i := ( 1! m! m c = 1 c i i =i c i where rus through all permutatios of δ(i We prove that for sufficietly small r we have (31 mi {c sup T (i c i O(log m r 1 sup T (i i A a r i A a i A a r r The left iequality is clear To see the right iequality we have from Lemma 35 that i C log r 1 for ay i A a r Whe i rus through A a r the umber of distict vectors ( δ(i 1, δ(i 2,, δ(i m is bouded by (C log r 1 m = O(log m r 1 Also there are at most m choices for the last letter of i The right iequality i (31 the follows Now for ay i = i 1 i i +1 A a r set t = Lemma 34, log T (i = O the other had we have a i = m log r ( t log t + t log c + O ( log a = a i +1 < r m where = δ(i ad = δ(i By a ( log t log a + O Hece Combiig the two estimates yields m log T (i ( t log t + t log c ( log (32 = log r m t + O log a

11 A CLASS OF SELF-AFFINE SETS AND SELF-AFFINE MEASURES 11 The coditio a i b i is equivalet to (33 t log b + b i m +1 t log a + a i +1 The propositio follows from (32 ad (33, by lettig teds to We ow prove Propositio 32 We will eed to ivoke the followig Large Deviatio Priciple: Lemma 36 (Large Deviatio Priciple Let p = (p 1,, p m be a probability vector For ay ε > 0 there exists a ω = ω(ε > 0 such that i B (ε p i < e ω for all sufficietly large, where (34 B (ε := with Σ = Σ(m { i Σ : i p > ε Proof Stadard The reader may see [3, Theorem 2110] for a proof Proof of Propositio 32 As with Propositio 31, we prove (i oly The others are proved usig idetical argumets Assume that p (log b log a = m p (log b log a = δ 0 > 0 For ay η > 0 let ε = ε(η = η/m where M = 2m ( log c + log a + log b The it is easily verified that for ay i Σ \ B (ε we have 1 log c i p log c < η, as well as Therefore (35 1 log a i p log a < η, 1 log b i log c i log a i = p log b < η m p log c m p log a + O(η Note that for this ε > 0 there is a ω = ω(ε > 0 such that i B (ε p i < e ω for all 0

12 12 DE-JUN FENG AND YANG WANG By Lemma 34, for ay i A r we have C 1 log r 1 i C log r 1 Let r > 0 be sufficietly small so that C 1 log r 1 0 We ow decompose A r ito A r,1 ad A r,2 with A r,1 = A r \ A r,2, ad A r,2 = A r ( B (ε By observig that log c i log r C 0 := C max 1 m { log c we obtai p i log c i C 0 p i C 0 e kω log r i A r,2 i A r,2 C 1 log r 1 k C log r 1 Hece p i log c i i A r,2 log r teds to 0 as r 0+ O the other had, because a i < r a δ(i we have log r = log a i + O(1 If i A r,1 the by (35 p i log c i log r i A r,1 = i A r,1 p i = p logc p log a 1 ( p logc p log a + O(η i A r,1 p i + O(η Sice i A r p i = 1 because {[i] : i A r is a partitio of Σ N ad p i e kn(ε 0 i A r,2 C 1 log r 1 k C log r 1 as r 0, we must have lim r 0 i A r,1 p i = 1 Now because m p (log b log a = δ 0 > 0, A r,1 A a r wheever η (ad hece ε is sufficietly small It follows that p i log c i (36 log r p i log c i log r p i log c i log r i A r,1 i A r i A a r Takig limit r 0 yields Ω a = p logc p log a To see that Ωb = 0 we oly eed to observe that by (36, Ω a + Ω b p i log c i = lim = Ω a r 0 log r i A r 4 Proof of Theorem 21 We adopt the followig defiitio from [26]: Defiitio 41 Let K be a compact set i R d Fix M, ε > 0 ad N N A coverig {G i i=1 of K by Borel sets is said to be (M, ε, N good if diam (G i Mε for all i, ad ay ε-cube i R d itersects at most N elemets i the coverig

13 A CLASS OF SELF-AFFINE SETS AND SELF-AFFINE MEASURES 13 Lemma 41 Let M, q > 0 ad N, d N There exists a costat C 1 = C 1 (M, N, d, q such that for ay compactly supported probability measure ν o R d ad ay (M, 2, N-good Borel coverig {G i of supp (ν we have C 1 1 τ (ν, q i (ν(g i q C 1 τ (ν, q where τ (ν, q is defied i (13 Proof See [26], Lemma 22 Let {φ m be a IFS i Rd For ay i = i 1 i 2 i Σ, Σ = {1, 2,, m we let φ i deote φ i1 φ i2 φ i Lemma 42 Let I = {φ m be a IFS i Rd ad p = (p 1, p 2,, p m be a probability vector The for ay compact set F we have ν I,p (F = lim p i i B where B = {i Σ : φ i (K F ad K = K(I Proof Stadard Lemma 43 Uder the assumptios of Theorem 21, for ay i Σ we have ν(φ i (K = p i, where ν := ν I,p ad K = K(I Proof Let T be the ope rectagle for the ROSC, so {φ (T m are disoit ope rectagles i T First we cosider the case i which φ (T T for some 1 m Without loss of geerality we assume that φ 1 (T T Let l Σ By Lemma 42 we have { B k = i Σ k : φ i (K φ l (K {l Σ k Hece ν(φ l (K = lim k i B k p i p l We prove the coverse I fact we prove ν(φ l (T p l Let ad set C k = {i Σ k : φ i (K φ l (T C 1 k = {i C k : φ i (K φ l (T, C 2 k = C k \ C 1 k

14 14 DE-JUN FENG AND YANG WANG Note that Ck 1 = {l Σk Hece lim k i Ck 1 p i = p l O the other had, Ck {i 2 = i 1 i 2 i k : i 1 i 2 i l, i 1 for > Hece i Ck 2 p i < (1 p 1 k 0 as k Thus lim k i C k p i = p l It follows that (41 ν(φ l (K ν(φ l (T = p l By cosiderig the iteratios of the IFS I it is clear that the above proof exteds to the case i which there exists a i Σ such that φ i (T T It remais to prove the lemma whe φ i (T T for all oempty i Σ I this case it is clear that K is cotaied i a lie parallel to oe of the axes, say, the horizotal axis The ν is idetical to its proectio ν x oto the x-axis up to a traslatio Furthermore the proectio IFS I x must satisfy the OSC (ad it is self-similar Therefore the lemma still holds Propositio 44 Let ν be a self-similar probability measure i R with supp (ν [c, d] For ay q, δ > 0 there exist costats C 1, C 2 > 0 depedig o ν, q, δ such that for ay > 0 we have C 1 τ(ν,q δ where I i = [c + (i 1(d c, c + i(d c ] (ν(i i q C 2 τ(ν,q+δ i=1 Proof It is kow that for a self-similar measure the L q spectrum exists, see Peres ad Solomyak [26] Set = d c, which is the legth of each iterval I i By defiitio ad Lemma 41 we have lim log i=0 (ν(i i q log = τ(ν, q Thus for ay δ > 0 there exists a 0 such that for all > 0 we have τ(ν,q δ (ν(i i q τ(ν,q+δ i=0 Now for 1 0 we simply choose C 1 ad C 2 to satisfy the iequalities of the propositio Proof of Theorem 21 Let r > 0 be sufficietly small We costruct a coverig {G i of supp (ν as follows: Let T be the ope rectagle associated with the ROSC for the IFS I Without loss of geerality we may assume that T is a uit square For a = (a m ad

15 A CLASS OF SELF-AFFINE SETS AND SELF-AFFINE MEASURES 15 b = (b m defie the subsets of A r, A a r ad A b r of Σ with Σ = {1, 2,, m as i Sectio 3 For ay i A a r by defiitio b i a i, ad we set w a (i := [b i /a i ] Note that φ i (T is a closed rectagle of width a i ad height b i ; so w a (i is the aspect ratio of the rectagle rouded off to a iteger We ow cut φ i (T horizotally ito w a (i equal rectagles of width a i ad height b i /w a (i Call these smaller rectagles {Ri,k a wa(i k=1 Similarly for ay i A b r by defiitio a i > b i, ad we set w b (i := [a i /b i ] φ i (T is a closed rectagle of width a i ad height b i We ow cut φ i (T vertically ito w b (i equal rectagles of width a i /w b (i ad height b i Call these smaller rectagles {R b i,k w b(i k=1 Observe that if s = mi{a, b ad s + = max{a, b the each Ri,k a ad Rb i,k has width ad height betwee s r ad r/s + Furthermore { C r = Ri,k a : i Aa r, 1 k w a (i { Ri,k b : i Ab r, 1 k w b (i is a coverig of supp (ν It follows that C r is a (M, r, N-good coverig of supp (ν with M = s /2 ad N = 4 The key is to estimate w a(i k=1 (ν(ra i,k q for i A a r ad w b (i k=1 (ν(rb i,k q for i A b r We make the followig claim: Claim: For ay δ > 0 there exist costats C 1 ad C 2 idepedet of i ad r such that (42 C 1 p q i (w a(i τ(νy,q δ w a(i (ν(ri,k a q C 2 p q i (w a(i τ(νy,q+δ k=1 Proof of Claim: The combiatio of Lemma 42 ad Lemma 43 implies that ν(ri,k {p a = lim : Σ, φ (K φ i (T, φ (K R a i,k = lim {p : {i Σ i, φ (K R a i,k = p i lim {p : Σ, φ i φ (K R a i,k But observe that the set { Σ : φ i φ (K Ri,k a is precisely the set { (43 Σ : φ y (K y I i,k, where φ y := π y φ π 1 y, K y = π y (K ad I i,k := [ c+(k 1(d c w a(i (So by assumptio actually d c = 1 Propositio 44 ow asserts that C 1 p q i (w a(i τ(νy,q δ w a(i, c+k(d c w a(i ] with [c, d] = π y (T (ν(ri,k a q C 2 p q i (w a(i τ(νy,q+δ k=1

16 16 DE-JUN FENG AND YANG WANG for some costats C 1 ad C 2, provig the claim By a idetical argumet we also have costats C 1 ad C 2 such that (44 C 1p q i (w b(i τ(νx,q δ for ay i A b r w b (i (ν(ri,k b q C 2p q i (w b(i τ(νx,q+δ k=1 To complete the proof of our theorem, D C r (ν(d q = It follows from (42 ad (44 that i A a r i A a r w a(i k=1 w b (i (ν(ri,k a q + (ν(ri,k b q i A a r w a(i i A b r k=1 C 1 p q i (w a(i τ(νy,q δ (ν(ri,k a q C 2 p q i (w a(i τ(νy,q+δ, ad similarly C 1 Note that i A b r k=1 w b (i i A b r k=1 i A a r p q i (w b(i τ(νx,q δ (ν(ri,k b q C 2p q i (w b(i τ(νx,q+δ b i 2a i i A b r w a (i b i a i Applyig Propositio 31 twice with c = {c m c = p q ( b a τ(νy,q δ ad c = p q ( b a τ(νy,q+δ respectively, ad with δ 0, yields wa(i log( i A lim a r k=1 (ν(ra i,k q = if r 0 log r t Γ(e a where e a := log b log a Similarly log i A lim b r r 0 wb (i k=1 (ν(rb i,k q log r t ( log t τ(ν y, qe a + q log p t log a ( t log t τ(ν x, qe b + q log p = if t Γ(e b t log b set to be where e b := log a log b The proof is fially complete by observig that for ay A B > 0 we have log A < log(a + B log A + log 2,, Proof of Theorem 22 It is clear from the proof of Theorem 21 that if all a b the τ(ν, q = θ a where θ a is give i Theorem 21 I this case Γ 0 := Γ(log b log b = {t =

17 A CLASS OF SELF-AFFINE SETS AND SELF-AFFINE MEASURES 17 (t 1,, t m : t 0 ad m t = 1 Hece m t ( log t τ(ν y, q log b + q log p (45 τ(ν, q = τ(ν y, q + if t Γ 0 m t log a We first simplify the otatio Set A = 1/a ad B = p q b τ(νy,q (46 τ(ν y, q τ(ν, q = sup t Γ 0 Let θ be the uique real root of the equatio m m t log B t m t log A A θ from the fact that A > 1 ad B > 0 for all The m t log B t m t log A θ = m t log A θ B t m t log A The (45 becomes B = 1 (the existece of θ follows Note that f(x = log x is covex By Jese s Iequality, t log A θ B ( m A θ B ( m log t log A θ B = 0 t t The = i the first iequality is achieved whe t = A θ B for 1 m It follows τ(ν y, q τ(ν, q = θ This proves the theorem 5 Proof of Theorem 25 Let ν be a compactly supported probability measure i R d For ay ε > 0 ad N N a family of Borel sets {G i is called a (ε, N-good partitio coverig of supp (ν if the followig coditios are met: (i i G i supp (ν ad ν(g i G = 0 for all i (ii Ay cube of side ε itersects at most N elemets of {G i ad diam (G i < ε Set f(x = x log(1/x = x log x ad defie h (ν = Q D f(ν(q, where D is the stadard partitio of R d by cubes of sides 2 defied i Sectio 1 We have Lemma 51 Let {G i be a (2, N-good partitio coverig of supp (ν where ν is ay compactly supported probability measure i R d The f(ν(g i h (ν C i where C = max (log N, log 2 d

18 18 DE-JUN FENG AND YANG WANG Proof It is easy to check that for all x 1,, x k 0 we have (51 f( k i=1 x i k i=1 f(x i f( k i=1 x i + ( k i=1 x i log k Write D = {Q ad cosider the refiemet G = {G i Q of the (2, N-good partitio coverig {G i Note that diam (G i < 2 implies that G i itersects at most 2 d cubes i D It follows from (51 that (52 f(ν(g i f(ν(g i Q f(ν(g i + log(2 d ν(g i Coversely, also by (51 we have (53 f(ν(q i f(ν(g i Q f(ν(q + log(n ν(q Summig up (52 over i yields (54 f(ν(g i i i, f(ν(g i Q i f(ν(g i + log(2 d, ad summig up (53 over yields (55 h (ν i, f(ν(g i Q h (ν + log N The lemma ow follows by combiig (54 ad (55 Propositio 52 Let ν be a self-similar probability measure i R with supp (ν [c, d] For ay δ > 0 there exist costats C 1, C 2 > 0 depedig o ν ad δ such that for ay > 0 we have (h(ν δ log + C 1 f(i i (h(ν + δ log + C 2, i=1 where I i = [c + (i 1(d c, c + i(d c Proof The propositio ca be proved by usig Lemma 51 ad a argumet idetical to that of Propositio 44 Proof of Theorem 25 For ν = ν I,p we may assume without loss of geerality that ν(l = 0 for ay lie L i R 2, for otherwise ν is i essece a self-similar measure i the oe dimesio, leavig us with othig to prove

19 A CLASS OF SELF-AFFINE SETS AND SELF-AFFINE MEASURES 19 We adopt the same otatios from the proof of Theorem 21 For ay r > 0 sufficietly small let C r be the coverig of supp (ν give by { { C r = Ri,k a : i Aa r, 1 k w a (i Ri,k b : i Ab r, 1 k w b (i as i the proof of Theorem 21 It follows that C r is a (r, N-good partitio coverig of K = supp (ν for some suitable N idepedet of r f(x = x log(1/x = x log x We first estimate w a(i k=1 f(ν(ra i,k for ay i Aa r By (43 we have ν(ri,k a = p { i lim Σ : φ y (K y I i,k = p i ν y (I i,k where K y ad I i,k are as i the proof of Theorem 21 Thus w a(i k=1 f(ν(r a i,k = p i w a(i k=1 We estimate Q C r f(ν(q for f(ν y (I i,k p i log p i It ow follows from Propositio 52 that for ay δ > 0 there are C 1 ad C 2 idepedet of r such that (56 p i (h(ν y δ log w a (i p i log p i +C 1 p i w a(i k=1 f(ν(r a i,k p i(h(ν y +δ log w a (i p i log p i +C 2 p i Similarly for ay i A b r there exist C 1 ad C 2 idepedet of r such that (57 p i (h(ν x δ log w b (i p i log p i +C 1p i Now, observe that b i 2a i w b (i k=1 f(ν(r b i,k p i(h(ν x +δ log w b (i p i log p i +C 2p i w a (i b i a i So log(b i /a i log 2 log w a (i log(b i /a i This meas we may replace w a (i i (56 with b i /a i, with oly the costats C 1 ad C 2 modified Now for (56 we apply Propositio 32 twice, with c = p 1 (b /a h(νy δ ad c = p 1 (b /a h(νy +δ respectively, ad set δ 0 Set e a := log b log a ad e a := log a log b It follows that (58 lim r 0 wa(i i A a r k=1 f(ν(ra i,k log r = h(νy p e a p log p, p log a

20 20 DE-JUN FENG AND YANG WANG if m p (log b log a > 0, ad 0 if m (59 lim r 0 i A b r wb (i k=1 f(ν(rb i,k log r p (log a log b > 0 Similarly, = h(νx p e b p log p, p log b if m p (log a log b > 0, ad 0 if m p (log b log a > 0 Combiig (58 ad (59 yields h(ν G C (510 lim r f(ν(g y p e a p log p, if p e a > 0, p log a = r 0 log r h(ν x p e b p log p, if p e b > 0 p log b To estimate the left-had side of the above equatio whe m p (log a log b = 0, by summig (56 over i A a r ad (57 over i A b r we have W r,1 f(ν(g W r,2 G C r with ad W r,1 := C 3 + p i (h(ν y δ log(b i /a i + p i (h(ν x δ log(a i /b i i A a r W 2,r := C 4 + p i (h(ν y + δ log(b i /a i + p i (h(ν x + δ log(a i /b i i A a r i A b r i A b r i A r p i log p i i A r p i log p i, where C 3 ad C 4 are costats idepedet of r Applyig Propositio 32 (iii we have G C (511 lim r f(ν(g p log p = r 0 log r p log a wheever p e a = 0 Note that C r is a (r, N-good partitio coverig of supp (ν for some costat N idepedet of r Takig r = 2 ad applyig Lemma 51 we have h (ν h(ν = lim log 2 = lim G C f(ν(g 2 G C = lim r f(ν(g log 2 r 0 log r The theorem ow follows by combiig (510 ad (511 Ackowledgemet The authors thak Dr Hui Rao for some helpful discussios about the box-dimesio of self-affie sets Refereces [1] T Bedford, Crikly curves, Markov partitios ad box dimesio i self-similar sets, PhD Thesis, Uiversity of Warwick, 1984 [2] R Cawley ad R D Mauldi, Multifractal decompositios of Mora fractals Adv Math 92 (1992,

21 A CLASS OF SELF-AFFINE SETS AND SELF-AFFINE MEASURES 21 [3] A Dembo ad O Zeitoui, Large deviatios techiques ad applicatios Secod editio Spriger- Verlag, New York, 1998 [4] K J Falcoer, The Hausdorff dimesio of self-affie fractals Math Proc Cambridge Philos Soc 103 (1988, o 2, [5] K J Falcoer, Fractal geometry Mathematical foudatios ad applicatios Joh Wiley & Sos, Ltd, Chichester, 1990 [6] K J Falcoer, Geeralized dimesios of measures o self-affie sets Noliearity 12 (1999, [7] A H Fa, K S Lau ad S M Ngai, Iterated fuctio systems with overlaps Asia J Math 4 (2000, [8] D J Feg, The limited Rademacher fuctios ad Beroulli covolutios associated with Pisot umbers Preprit [9] D J Feg, Smoothess of the L q -spectrum of self-similar measures with overlaps J Lodo Math Soc (2 68 (2003, [10] H Y Hu, Box dimesios ad topological pressure for some expadig maps Comm Math Phys 191 (1998, [11] S Hua ad H Rao, Graph-directed structures of self-similar sets with overlaps Chiese A Math Ser B 21 (2000, [12] I Hueter ad S P Lalley, Falcoer s formula for the Hausdorff dimesio of a self-affie set i R 2 Ergodic Theory Dyam Systems 15 (1995, [13] J E Hutchiso, Fractals ad self-similarity Idiaa Uiv Math J 30 (1981, [14] R Keyo ad Y Peres, Hausdorff dimesios of sofic affie-ivariat sets Israel J Math 94 (1996, [15] J F Kig, The sigularity spectrum for geeral Sierpiński carpets Adv Math 116 (1995, 1 11 [16] S P Lalley, β-expasios with deleted digits for Pisot umbers β Tras Amer Math Soc 349 (1997, [17] S P Lalley ad D Gatzouras, Hausdorff ad box dimesios of certai self-affie fractals Idiaa Uiv Math J 41 (1992, [18] K S Lau ad S M Ngai, L q -spectrum of the Beroulli covolutio associated with the golde ratio Studia Math 131 (1998, o 3, [19] K S Lau ad X Y Wag, Some exceptioal pheomea i multifractal formalism: Part I Preprit [20] F Ledrappier ad L S Youg, The metric etropy of diffeomorphisms II Relatios betwee etropy, expoets ad dimesio A of Math (2 122 (1985, [21] C McMulle, The Hausdorff dimesio of geeral Sierpiński carpets Nagoya Math J 96 (1984, 1 9 [22] S M Ngai, A dimesio result arisig from the L q -spectrum of a measure Proc Amer Math Soc 125 (1997, [23] S M Ngai ad Y Wag, Hausdorff dimesio of self-similar sets with overlaps J Lodo Math Soc (2 63 (2001, [24] L Olse, A multifractal formalism Adv Math 116 (1995, [25] L Olse, Self-affie multifractal Sierpiski spoges i R d Pacific J Math 183 (1998, [26] Y Peres ad B Solomyak, Existece of L q dimesios ad etropy dimesio for self-coformal measures Idiaa Uiv Math J 49 (2000, [27] B Solomyak, Measure ad dimesio for some fractal families Math Proc Cambridge Philos Soc 124 (1998, o 3, Departmet of Mathematical Scieces, Tsighua Uiversity, Beiig, , P R Chia address: dfeg@mathtsighuaeduc School of Mathematics, Georgia Istitute of Techology, Atlata, Georgia 30332, USA address: wag@mathgatechedu

Convexity, Inequalities, and Norms

Convexity, Inequalities, and Norms Covexity, Iequalities, ad Norms Covex Fuctios You are probably familiar with the otio of cocavity of fuctios. Give a twicedifferetiable fuctio ϕ: R R, We say that ϕ is covex (or cocave up) if ϕ (x) 0 for

More information

Department of Computer Science, University of Otago

Department of Computer Science, University of Otago Departmet of Computer Sciece, Uiversity of Otago Techical Report OUCS-2006-09 Permutatios Cotaiig May Patters Authors: M.H. Albert Departmet of Computer Sciece, Uiversity of Otago Micah Colema, Rya Fly

More information

SAMPLE QUESTIONS FOR FINAL EXAM. (1) (2) (3) (4) Find the following using the definition of the Riemann integral: (2x + 1)dx

SAMPLE QUESTIONS FOR FINAL EXAM. (1) (2) (3) (4) Find the following using the definition of the Riemann integral: (2x + 1)dx SAMPLE QUESTIONS FOR FINAL EXAM REAL ANALYSIS I FALL 006 3 4 Fid the followig usig the defiitio of the Riema itegral: a 0 x + dx 3 Cosider the partitio P x 0 3, x 3 +, x 3 +,......, x 3 3 + 3 of the iterval

More information

Asymptotic Growth of Functions

Asymptotic Growth of Functions CMPS Itroductio to Aalysis of Algorithms Fall 3 Asymptotic Growth of Fuctios We itroduce several types of asymptotic otatio which are used to compare the performace ad efficiecy of algorithms As we ll

More information

Lecture 13. Lecturer: Jonathan Kelner Scribe: Jonathan Pines (2009)

Lecture 13. Lecturer: Jonathan Kelner Scribe: Jonathan Pines (2009) 18.409 A Algorithmist s Toolkit October 27, 2009 Lecture 13 Lecturer: Joatha Keler Scribe: Joatha Pies (2009) 1 Outlie Last time, we proved the Bru-Mikowski iequality for boxes. Today we ll go over the

More information

In nite Sequences. Dr. Philippe B. Laval Kennesaw State University. October 9, 2008

In nite Sequences. Dr. Philippe B. Laval Kennesaw State University. October 9, 2008 I ite Sequeces Dr. Philippe B. Laval Keesaw State Uiversity October 9, 2008 Abstract This had out is a itroductio to i ite sequeces. mai de itios ad presets some elemetary results. It gives the I ite Sequeces

More information

Properties of MLE: consistency, asymptotic normality. Fisher information.

Properties of MLE: consistency, asymptotic normality. Fisher information. Lecture 3 Properties of MLE: cosistecy, asymptotic ormality. Fisher iformatio. I this sectio we will try to uderstad why MLEs are good. Let us recall two facts from probability that we be used ofte throughout

More information

Lecture 4: Cauchy sequences, Bolzano-Weierstrass, and the Squeeze theorem

Lecture 4: Cauchy sequences, Bolzano-Weierstrass, and the Squeeze theorem Lecture 4: Cauchy sequeces, Bolzao-Weierstrass, ad the Squeeze theorem The purpose of this lecture is more modest tha the previous oes. It is to state certai coditios uder which we are guarateed that limits

More information

THE ABRACADABRA PROBLEM

THE ABRACADABRA PROBLEM THE ABRACADABRA PROBLEM FRANCESCO CARAVENNA Abstract. We preset a detailed solutio of Exercise E0.6 i [Wil9]: i a radom sequece of letters, draw idepedetly ad uiformly from the Eglish alphabet, the expected

More information

Chapter 5: Inner Product Spaces

Chapter 5: Inner Product Spaces Chapter 5: Ier Product Spaces Chapter 5: Ier Product Spaces SECION A Itroductio to Ier Product Spaces By the ed of this sectio you will be able to uderstad what is meat by a ier product space give examples

More information

A probabilistic proof of a binomial identity

A probabilistic proof of a binomial identity A probabilistic proof of a biomial idetity Joatho Peterso Abstract We give a elemetary probabilistic proof of a biomial idetity. The proof is obtaied by computig the probability of a certai evet i two

More information

I. Chi-squared Distributions

I. Chi-squared Distributions 1 M 358K Supplemet to Chapter 23: CHI-SQUARED DISTRIBUTIONS, T-DISTRIBUTIONS, AND DEGREES OF FREEDOM To uderstad t-distributios, we first eed to look at aother family of distributios, the chi-squared distributios.

More information

Approximating Area under a curve with rectangles. To find the area under a curve we approximate the area using rectangles and then use limits to find

Approximating Area under a curve with rectangles. To find the area under a curve we approximate the area using rectangles and then use limits to find 1.8 Approximatig Area uder a curve with rectagles 1.6 To fid the area uder a curve we approximate the area usig rectagles ad the use limits to fid 1.4 the area. Example 1 Suppose we wat to estimate 1.

More information

Irreducible polynomials with consecutive zero coefficients

Irreducible polynomials with consecutive zero coefficients Irreducible polyomials with cosecutive zero coefficiets Theodoulos Garefalakis Departmet of Mathematics, Uiversity of Crete, 71409 Heraklio, Greece Abstract Let q be a prime power. We cosider the problem

More information

Theorems About Power Series

Theorems About Power Series Physics 6A Witer 20 Theorems About Power Series Cosider a power series, f(x) = a x, () where the a are real coefficiets ad x is a real variable. There exists a real o-egative umber R, called the radius

More information

SECTION 1.5 : SUMMATION NOTATION + WORK WITH SEQUENCES

SECTION 1.5 : SUMMATION NOTATION + WORK WITH SEQUENCES SECTION 1.5 : SUMMATION NOTATION + WORK WITH SEQUENCES Read Sectio 1.5 (pages 5 9) Overview I Sectio 1.5 we lear to work with summatio otatio ad formulas. We will also itroduce a brief overview of sequeces,

More information

Lecture 5: Span, linear independence, bases, and dimension

Lecture 5: Span, linear independence, bases, and dimension Lecture 5: Spa, liear idepedece, bases, ad dimesio Travis Schedler Thurs, Sep 23, 2010 (versio: 9/21 9:55 PM) 1 Motivatio Motivatio To uderstad what it meas that R has dimesio oe, R 2 dimesio 2, etc.;

More information

Our aim is to show that under reasonable assumptions a given 2π-periodic function f can be represented as convergent series

Our aim is to show that under reasonable assumptions a given 2π-periodic function f can be represented as convergent series 8 Fourier Series Our aim is to show that uder reasoable assumptios a give -periodic fuctio f ca be represeted as coverget series f(x) = a + (a cos x + b si x). (8.) By defiitio, the covergece of the series

More information

Incremental calculation of weighted mean and variance

Incremental calculation of weighted mean and variance Icremetal calculatio of weighted mea ad variace Toy Fich faf@cam.ac.uk dot@dotat.at Uiversity of Cambridge Computig Service February 009 Abstract I these otes I eplai how to derive formulae for umerically

More information

THIN SEQUENCES AND THE GRAM MATRIX PAMELA GORKIN, JOHN E. MCCARTHY, SANDRA POTT, AND BRETT D. WICK

THIN SEQUENCES AND THE GRAM MATRIX PAMELA GORKIN, JOHN E. MCCARTHY, SANDRA POTT, AND BRETT D. WICK THIN SEQUENCES AND THE GRAM MATRIX PAMELA GORKIN, JOHN E MCCARTHY, SANDRA POTT, AND BRETT D WICK Abstract We provide a ew proof of Volberg s Theorem characterizig thi iterpolatig sequeces as those for

More information

CS103X: Discrete Structures Homework 4 Solutions

CS103X: Discrete Structures Homework 4 Solutions CS103X: Discrete Structures Homewor 4 Solutios Due February 22, 2008 Exercise 1 10 poits. Silico Valley questios: a How may possible six-figure salaries i whole dollar amouts are there that cotai at least

More information

Chapter 6: Variance, the law of large numbers and the Monte-Carlo method

Chapter 6: Variance, the law of large numbers and the Monte-Carlo method Chapter 6: Variace, the law of large umbers ad the Mote-Carlo method Expected value, variace, ad Chebyshev iequality. If X is a radom variable recall that the expected value of X, E[X] is the average value

More information

A Faster Clause-Shortening Algorithm for SAT with No Restriction on Clause Length

A Faster Clause-Shortening Algorithm for SAT with No Restriction on Clause Length Joural o Satisfiability, Boolea Modelig ad Computatio 1 2005) 49-60 A Faster Clause-Shorteig Algorithm for SAT with No Restrictio o Clause Legth Evgey Datsi Alexader Wolpert Departmet of Computer Sciece

More information

Class Meeting # 16: The Fourier Transform on R n

Class Meeting # 16: The Fourier Transform on R n MATH 18.152 COUSE NOTES - CLASS MEETING # 16 18.152 Itroductio to PDEs, Fall 2011 Professor: Jared Speck Class Meetig # 16: The Fourier Trasform o 1. Itroductio to the Fourier Trasform Earlier i the course,

More information

Infinite Sequences and Series

Infinite Sequences and Series CHAPTER 4 Ifiite Sequeces ad Series 4.1. Sequeces A sequece is a ifiite ordered list of umbers, for example the sequece of odd positive itegers: 1, 3, 5, 7, 9, 11, 13, 15, 17, 19, 21, 23, 25, 27, 29...

More information

1. MATHEMATICAL INDUCTION

1. MATHEMATICAL INDUCTION 1. MATHEMATICAL INDUCTION EXAMPLE 1: Prove that for ay iteger 1. Proof: 1 + 2 + 3 +... + ( + 1 2 (1.1 STEP 1: For 1 (1.1 is true, sice 1 1(1 + 1. 2 STEP 2: Suppose (1.1 is true for some k 1, that is 1

More information

CME 302: NUMERICAL LINEAR ALGEBRA FALL 2005/06 LECTURE 8

CME 302: NUMERICAL LINEAR ALGEBRA FALL 2005/06 LECTURE 8 CME 30: NUMERICAL LINEAR ALGEBRA FALL 005/06 LECTURE 8 GENE H GOLUB 1 Positive Defiite Matrices A matrix A is positive defiite if x Ax > 0 for all ozero x A positive defiite matrix has real ad positive

More information

Example 2 Find the square root of 0. The only square root of 0 is 0 (since 0 is not positive or negative, so those choices don t exist here).

Example 2 Find the square root of 0. The only square root of 0 is 0 (since 0 is not positive or negative, so those choices don t exist here). BEGINNING ALGEBRA Roots ad Radicals (revised summer, 00 Olso) Packet to Supplemet the Curret Textbook - Part Review of Square Roots & Irratioals (This portio ca be ay time before Part ad should mostly

More information

Trigonometric Form of a Complex Number. The Complex Plane. axis. ( 2, 1) or 2 i FIGURE 6.44. The absolute value of the complex number z a bi is

Trigonometric Form of a Complex Number. The Complex Plane. axis. ( 2, 1) or 2 i FIGURE 6.44. The absolute value of the complex number z a bi is 0_0605.qxd /5/05 0:45 AM Page 470 470 Chapter 6 Additioal Topics i Trigoometry 6.5 Trigoometric Form of a Complex Number What you should lear Plot complex umbers i the complex plae ad fid absolute values

More information

Chapter 7 Methods of Finding Estimators

Chapter 7 Methods of Finding Estimators Chapter 7 for BST 695: Special Topics i Statistical Theory. Kui Zhag, 011 Chapter 7 Methods of Fidig Estimators Sectio 7.1 Itroductio Defiitio 7.1.1 A poit estimator is ay fuctio W( X) W( X1, X,, X ) of

More information

Annuities Under Random Rates of Interest II By Abraham Zaks. Technion I.I.T. Haifa ISRAEL and Haifa University Haifa ISRAEL.

Annuities Under Random Rates of Interest II By Abraham Zaks. Technion I.I.T. Haifa ISRAEL and Haifa University Haifa ISRAEL. Auities Uder Radom Rates of Iterest II By Abraham Zas Techio I.I.T. Haifa ISRAEL ad Haifa Uiversity Haifa ISRAEL Departmet of Mathematics, Techio - Israel Istitute of Techology, 3000, Haifa, Israel I memory

More information

Section 11.3: The Integral Test

Section 11.3: The Integral Test Sectio.3: The Itegral Test Most of the series we have looked at have either diverged or have coverged ad we have bee able to fid what they coverge to. I geeral however, the problem is much more difficult

More information

FIBONACCI NUMBERS: AN APPLICATION OF LINEAR ALGEBRA. 1. Powers of a matrix

FIBONACCI NUMBERS: AN APPLICATION OF LINEAR ALGEBRA. 1. Powers of a matrix FIBONACCI NUMBERS: AN APPLICATION OF LINEAR ALGEBRA. Powers of a matrix We begi with a propositio which illustrates the usefuless of the diagoalizatio. Recall that a square matrix A is diogaalizable if

More information

Sequences and Series

Sequences and Series CHAPTER 9 Sequeces ad Series 9.. Covergece: Defiitio ad Examples Sequeces The purpose of this chapter is to itroduce a particular way of geeratig algorithms for fidig the values of fuctios defied by their

More information

Confidence Intervals for One Mean

Confidence Intervals for One Mean Chapter 420 Cofidece Itervals for Oe Mea Itroductio This routie calculates the sample size ecessary to achieve a specified distace from the mea to the cofidece limit(s) at a stated cofidece level for a

More information

Basic Elements of Arithmetic Sequences and Series

Basic Elements of Arithmetic Sequences and Series MA40S PRE-CALCULUS UNIT G GEOMETRIC SEQUENCES CLASS NOTES (COMPLETED NO NEED TO COPY NOTES FROM OVERHEAD) Basic Elemets of Arithmetic Sequeces ad Series Objective: To establish basic elemets of arithmetic

More information

Permutations, the Parity Theorem, and Determinants

Permutations, the Parity Theorem, and Determinants 1 Permutatios, the Parity Theorem, ad Determiats Joh A. Guber Departmet of Electrical ad Computer Egieerig Uiversity of Wiscosi Madiso Cotets 1 What is a Permutatio 1 2 Cycles 2 2.1 Traspositios 4 3 Orbits

More information

THE LEAST COMMON MULTIPLE OF A QUADRATIC SEQUENCE

THE LEAST COMMON MULTIPLE OF A QUADRATIC SEQUENCE THE LEAST COMMON MULTIPLE OF A QUADRATIC SEQUENCE JAVIER CILLERUELO Abstract. We obtai, for ay irreducible quadratic olyomial f(x = ax 2 + bx + c, the asymtotic estimate log l.c.m. {f(1,..., f(} log. Whe

More information

Factors of sums of powers of binomial coefficients

Factors of sums of powers of binomial coefficients ACTA ARITHMETICA LXXXVI.1 (1998) Factors of sums of powers of biomial coefficiets by Neil J. Cali (Clemso, S.C.) Dedicated to the memory of Paul Erdős 1. Itroductio. It is well ow that if ( ) a f,a = the

More information

MARTINGALES AND A BASIC APPLICATION

MARTINGALES AND A BASIC APPLICATION MARTINGALES AND A BASIC APPLICATION TURNER SMITH Abstract. This paper will develop the measure-theoretic approach to probability i order to preset the defiitio of martigales. From there we will apply this

More information

On the L p -conjecture for locally compact groups

On the L p -conjecture for locally compact groups Arch. Math. 89 (2007), 237 242 c 2007 Birkhäuser Verlag Basel/Switzerlad 0003/889X/030237-6, ublished olie 2007-08-0 DOI 0.007/s0003-007-993-x Archiv der Mathematik O the L -cojecture for locally comact

More information

THE HEIGHT OF q-binary SEARCH TREES

THE HEIGHT OF q-binary SEARCH TREES THE HEIGHT OF q-binary SEARCH TREES MICHAEL DRMOTA AND HELMUT PRODINGER Abstract. q biary search trees are obtaied from words, equipped with the geometric distributio istead of permutatios. The average

More information

WHEN IS THE (CO)SINE OF A RATIONAL ANGLE EQUAL TO A RATIONAL NUMBER?

WHEN IS THE (CO)SINE OF A RATIONAL ANGLE EQUAL TO A RATIONAL NUMBER? WHEN IS THE (CO)SINE OF A RATIONAL ANGLE EQUAL TO A RATIONAL NUMBER? JÖRG JAHNEL 1. My Motivatio Some Sort of a Itroductio Last term I tought Topological Groups at the Göttige Georg August Uiversity. This

More information

Lecture 4: Cheeger s Inequality

Lecture 4: Cheeger s Inequality Spectral Graph Theory ad Applicatios WS 0/0 Lecture 4: Cheeger s Iequality Lecturer: Thomas Sauerwald & He Su Statemet of Cheeger s Iequality I this lecture we assume for simplicity that G is a d-regular

More information

Solutions to Exercises Chapter 4: Recurrence relations and generating functions

Solutions to Exercises Chapter 4: Recurrence relations and generating functions Solutios to Exercises Chapter 4: Recurrece relatios ad geeratig fuctios 1 (a) There are seatig positios arraged i a lie. Prove that the umber of ways of choosig a subset of these positios, with o two chose

More information

Discrete Mathematics and Probability Theory Spring 2014 Anant Sahai Note 13

Discrete Mathematics and Probability Theory Spring 2014 Anant Sahai Note 13 EECS 70 Discrete Mathematics ad Probability Theory Sprig 2014 Aat Sahai Note 13 Itroductio At this poit, we have see eough examples that it is worth just takig stock of our model of probability ad may

More information

Metric, Normed, and Topological Spaces

Metric, Normed, and Topological Spaces Chapter 13 Metric, Normed, ad Topological Spaces A metric space is a set X that has a otio of the distace d(x, y) betwee every pair of poits x, y X. A fudametal example is R with the absolute-value metric

More information

Soving Recurrence Relations

Soving Recurrence Relations Sovig Recurrece Relatios Part 1. Homogeeous liear 2d degree relatios with costat coefficiets. Cosider the recurrece relatio ( ) T () + at ( 1) + bt ( 2) = 0 This is called a homogeeous liear 2d degree

More information

Section 8.3 : De Moivre s Theorem and Applications

Section 8.3 : De Moivre s Theorem and Applications The Sectio 8 : De Moivre s Theorem ad Applicatios Let z 1 ad z be complex umbers, where z 1 = r 1, z = r, arg(z 1 ) = θ 1, arg(z ) = θ z 1 = r 1 (cos θ 1 + i si θ 1 ) z = r (cos θ + i si θ ) ad z 1 z =

More information

Definition. A variable X that takes on values X 1, X 2, X 3,...X k with respective frequencies f 1, f 2, f 3,...f k has mean

Definition. A variable X that takes on values X 1, X 2, X 3,...X k with respective frequencies f 1, f 2, f 3,...f k has mean 1 Social Studies 201 October 13, 2004 Note: The examples i these otes may be differet tha used i class. However, the examples are similar ad the methods used are idetical to what was preseted i class.

More information

4.3. The Integral and Comparison Tests

4.3. The Integral and Comparison Tests 4.3. THE INTEGRAL AND COMPARISON TESTS 9 4.3. The Itegral ad Compariso Tests 4.3.. The Itegral Test. Suppose f is a cotiuous, positive, decreasig fuctio o [, ), ad let a = f(). The the covergece or divergece

More information

1 Computing the Standard Deviation of Sample Means

1 Computing the Standard Deviation of Sample Means Computig the Stadard Deviatio of Sample Meas Quality cotrol charts are based o sample meas ot o idividual values withi a sample. A sample is a group of items, which are cosidered all together for our aalysis.

More information

Vladimir N. Burkov, Dmitri A. Novikov MODELS AND METHODS OF MULTIPROJECTS MANAGEMENT

Vladimir N. Burkov, Dmitri A. Novikov MODELS AND METHODS OF MULTIPROJECTS MANAGEMENT Keywords: project maagemet, resource allocatio, etwork plaig Vladimir N Burkov, Dmitri A Novikov MODELS AND METHODS OF MULTIPROJECTS MANAGEMENT The paper deals with the problems of resource allocatio betwee

More information

Notes on exponential generating functions and structures.

Notes on exponential generating functions and structures. Notes o expoetial geeratig fuctios ad structures. 1. The cocept of a structure. Cosider the followig coutig problems: (1) to fid for each the umber of partitios of a -elemet set, (2) to fid for each the

More information

Entropy of bi-capacities

Entropy of bi-capacities Etropy of bi-capacities Iva Kojadiovic LINA CNRS FRE 2729 Site école polytechique de l uiv. de Nates Rue Christia Pauc 44306 Nates, Frace iva.kojadiovic@uiv-ates.fr Jea-Luc Marichal Applied Mathematics

More information

3. Greatest Common Divisor - Least Common Multiple

3. Greatest Common Divisor - Least Common Multiple 3 Greatest Commo Divisor - Least Commo Multiple Defiitio 31: The greatest commo divisor of two atural umbers a ad b is the largest atural umber c which divides both a ad b We deote the greatest commo gcd

More information

How To Solve An Old Japanese Geometry Problem

How To Solve An Old Japanese Geometry Problem 116 Taget circles i the ratio 2 : 1 Hiroshi Okumura ad Masayuki Wataabe I this article we cosider the followig old Japaese geometry problem (see Figure 1), whose statemet i [1, p. 39] is missig the coditio

More information

5.3. Generalized Permutations and Combinations

5.3. Generalized Permutations and Combinations 53 GENERALIZED PERMUTATIONS AND COMBINATIONS 73 53 Geeralized Permutatios ad Combiatios 53 Permutatios with Repeated Elemets Assume that we have a alphabet with letters ad we wat to write all possible

More information

The analysis of the Cournot oligopoly model considering the subjective motive in the strategy selection

The analysis of the Cournot oligopoly model considering the subjective motive in the strategy selection The aalysis of the Courot oligopoly model cosiderig the subjective motive i the strategy selectio Shigehito Furuyama Teruhisa Nakai Departmet of Systems Maagemet Egieerig Faculty of Egieerig Kasai Uiversity

More information

Week 3 Conditional probabilities, Bayes formula, WEEK 3 page 1 Expected value of a random variable

Week 3 Conditional probabilities, Bayes formula, WEEK 3 page 1 Expected value of a random variable Week 3 Coditioal probabilities, Bayes formula, WEEK 3 page 1 Expected value of a radom variable We recall our discussio of 5 card poker hads. Example 13 : a) What is the probability of evet A that a 5

More information

Normal Distribution.

Normal Distribution. Normal Distributio www.icrf.l Normal distributio I probability theory, the ormal or Gaussia distributio, is a cotiuous probability distributio that is ofte used as a first approimatio to describe realvalued

More information

A sharp Trudinger-Moser type inequality for unbounded domains in R n

A sharp Trudinger-Moser type inequality for unbounded domains in R n A sharp Trudiger-Moser type iequality for ubouded domais i R Yuxiag Li ad Berhard Ruf Abstract The Trudiger-Moser iequality states that for fuctios u H, 0 (Ω) (Ω R a bouded domai) with Ω u dx oe has Ω

More information

5 Boolean Decision Trees (February 11)

5 Boolean Decision Trees (February 11) 5 Boolea Decisio Trees (February 11) 5.1 Graph Coectivity Suppose we are give a udirected graph G, represeted as a boolea adjacecy matrix = (a ij ), where a ij = 1 if ad oly if vertices i ad j are coected

More information

CHAPTER 7: Central Limit Theorem: CLT for Averages (Means)

CHAPTER 7: Central Limit Theorem: CLT for Averages (Means) CHAPTER 7: Cetral Limit Theorem: CLT for Averages (Meas) X = the umber obtaied whe rollig oe six sided die oce. If we roll a six sided die oce, the mea of the probability distributio is X P(X = x) Simulatio:

More information

CS103A Handout 23 Winter 2002 February 22, 2002 Solving Recurrence Relations

CS103A Handout 23 Winter 2002 February 22, 2002 Solving Recurrence Relations CS3A Hadout 3 Witer 00 February, 00 Solvig Recurrece Relatios Itroductio A wide variety of recurrece problems occur i models. Some of these recurrece relatios ca be solved usig iteratio or some other ad

More information

Data Analysis and Statistical Behaviors of Stock Market Fluctuations

Data Analysis and Statistical Behaviors of Stock Market Fluctuations 44 JOURNAL OF COMPUTERS, VOL. 3, NO. 0, OCTOBER 2008 Data Aalysis ad Statistical Behaviors of Stock Market Fluctuatios Ju Wag Departmet of Mathematics, Beijig Jiaotog Uiversity, Beijig 00044, Chia Email:

More information

Institute of Actuaries of India Subject CT1 Financial Mathematics

Institute of Actuaries of India Subject CT1 Financial Mathematics Istitute of Actuaries of Idia Subject CT1 Fiacial Mathematics For 2014 Examiatios Subject CT1 Fiacial Mathematics Core Techical Aim The aim of the Fiacial Mathematics subject is to provide a groudig i

More information

A Guide to the Pricing Conventions of SFE Interest Rate Products

A Guide to the Pricing Conventions of SFE Interest Rate Products A Guide to the Pricig Covetios of SFE Iterest Rate Products SFE 30 Day Iterbak Cash Rate Futures Physical 90 Day Bak Bills SFE 90 Day Bak Bill Futures SFE 90 Day Bak Bill Futures Tick Value Calculatios

More information

Overview on S-Box Design Principles

Overview on S-Box Design Principles Overview o S-Box Desig Priciples Debdeep Mukhopadhyay Assistat Professor Departmet of Computer Sciece ad Egieerig Idia Istitute of Techology Kharagpur INDIA -721302 What is a S-Box? S-Boxes are Boolea

More information

. P. 4.3 Basic feasible solutions and vertices of polyhedra. x 1. x 2

. P. 4.3 Basic feasible solutions and vertices of polyhedra. x 1. x 2 4. Basic feasible solutios ad vertices of polyhedra Due to the fudametal theorem of Liear Programmig, to solve ay LP it suffices to cosider the vertices (fiitely may) of the polyhedro P of the feasible

More information

Output Analysis (2, Chapters 10 &11 Law)

Output Analysis (2, Chapters 10 &11 Law) B. Maddah ENMG 6 Simulatio 05/0/07 Output Aalysis (, Chapters 10 &11 Law) Comparig alterative system cofiguratio Sice the output of a simulatio is radom, the comparig differet systems via simulatio should

More information

INFINITE SERIES KEITH CONRAD

INFINITE SERIES KEITH CONRAD INFINITE SERIES KEITH CONRAD. Itroductio The two basic cocepts of calculus, differetiatio ad itegratio, are defied i terms of limits (Newto quotiets ad Riema sums). I additio to these is a third fudametal

More information

3 Basic Definitions of Probability Theory

3 Basic Definitions of Probability Theory 3 Basic Defiitios of Probability Theory 3defprob.tex: Feb 10, 2003 Classical probability Frequecy probability axiomatic probability Historical developemet: Classical Frequecy Axiomatic The Axiomatic defiitio

More information

0.7 0.6 0.2 0 0 96 96.5 97 97.5 98 98.5 99 99.5 100 100.5 96.5 97 97.5 98 98.5 99 99.5 100 100.5

0.7 0.6 0.2 0 0 96 96.5 97 97.5 98 98.5 99 99.5 100 100.5 96.5 97 97.5 98 98.5 99 99.5 100 100.5 Sectio 13 Kolmogorov-Smirov test. Suppose that we have a i.i.d. sample X 1,..., X with some ukow distributio P ad we would like to test the hypothesis that P is equal to a particular distributio P 0, i.e.

More information

Analysis Notes (only a draft, and the first one!)

Analysis Notes (only a draft, and the first one!) Aalysis Notes (oly a draft, ad the first oe!) Ali Nesi Mathematics Departmet Istabul Bilgi Uiversity Kuştepe Şişli Istabul Turkey aesi@bilgi.edu.tr Jue 22, 2004 2 Cotets 1 Prelimiaries 9 1.1 Biary Operatio...........................

More information

Convex Bodies of Minimal Volume, Surface Area and Mean Width with Respect to Thin Shells

Convex Bodies of Minimal Volume, Surface Area and Mean Width with Respect to Thin Shells Caad. J. Math. Vol. 60 (1), 2008 pp. 3 32 Covex Bodies of Miimal Volume, Surface Area ad Mea Width with Respect to Thi Shells Károly Böröczky, Károly J. Böröczky, Carste Schütt, ad Gergely Witsche Abstract.

More information

Lipschitz maps and nets in Euclidean space

Lipschitz maps and nets in Euclidean space Lipschitz maps ad ets i Euclidea space Curtis T. McMulle 1 April, 1997 1 Itroductio I this paper we discuss the followig three questios. 1. Give a real-valued fuctio f L (R ) with if f(x) > 0, is there

More information

ON THE DENSE TRAJECTORY OF LASOTA EQUATION

ON THE DENSE TRAJECTORY OF LASOTA EQUATION UNIVERSITATIS IAGELLONICAE ACTA MATHEMATICA, FASCICULUS XLIII 2005 ON THE DENSE TRAJECTORY OF LASOTA EQUATION by Atoi Leo Dawidowicz ad Najemedi Haribash Abstract. I preseted paper the dese trajectory

More information

A note on the boundary behavior for a modified Green function in the upper-half space

A note on the boundary behavior for a modified Green function in the upper-half space Zhag ad Pisarev Boudary Value Problems (015) 015:114 DOI 10.1186/s13661-015-0363-z RESEARCH Ope Access A ote o the boudary behavior for a modified Gree fuctio i the upper-half space Yulia Zhag1 ad Valery

More information

ON THE EDGE-BANDWIDTH OF GRAPH PRODUCTS

ON THE EDGE-BANDWIDTH OF GRAPH PRODUCTS ON THE EDGE-BANDWIDTH OF GRAPH PRODUCTS JÓZSEF BALOGH, DHRUV MUBAYI, AND ANDRÁS PLUHÁR Abstract The edge-badwidth of a graph G is the badwidth of the lie graph of G We show asymptotically tight bouds o

More information

THE ARITHMETIC OF INTEGERS. - multiplication, exponentiation, division, addition, and subtraction

THE ARITHMETIC OF INTEGERS. - multiplication, exponentiation, division, addition, and subtraction THE ARITHMETIC OF INTEGERS - multiplicatio, expoetiatio, divisio, additio, ad subtractio What to do ad what ot to do. THE INTEGERS Recall that a iteger is oe of the whole umbers, which may be either positive,

More information

NATIONAL SENIOR CERTIFICATE GRADE 12

NATIONAL SENIOR CERTIFICATE GRADE 12 NATIONAL SENIOR CERTIFICATE GRADE MATHEMATICS P EXEMPLAR 04 MARKS: 50 TIME: 3 hours This questio paper cosists of 8 pages ad iformatio sheet. Please tur over Mathematics/P DBE/04 NSC Grade Eemplar INSTRUCTIONS

More information

Modified Line Search Method for Global Optimization

Modified Line Search Method for Global Optimization Modified Lie Search Method for Global Optimizatio Cria Grosa ad Ajith Abraham Ceter of Excellece for Quatifiable Quality of Service Norwegia Uiversity of Sciece ad Techology Trodheim, Norway {cria, ajith}@q2s.tu.o

More information

BENEFIT-COST ANALYSIS Financial and Economic Appraisal using Spreadsheets

BENEFIT-COST ANALYSIS Financial and Economic Appraisal using Spreadsheets BENEIT-CST ANALYSIS iacial ad Ecoomic Appraisal usig Spreadsheets Ch. 2: Ivestmet Appraisal - Priciples Harry Campbell & Richard Brow School of Ecoomics The Uiversity of Queeslad Review of basic cocepts

More information

Rényi Divergence and L p -affine surface area for convex bodies

Rényi Divergence and L p -affine surface area for convex bodies Réyi Divergece ad L p -affie surface area for covex bodies Elisabeth M. Werer Abstract We show that the fudametal objects of the L p -Bru-Mikowski theory, amely the L p -affie surface areas for a covex

More information

The following example will help us understand The Sampling Distribution of the Mean. C1 C2 C3 C4 C5 50 miles 84 miles 38 miles 120 miles 48 miles

The following example will help us understand The Sampling Distribution of the Mean. C1 C2 C3 C4 C5 50 miles 84 miles 38 miles 120 miles 48 miles The followig eample will help us uderstad The Samplig Distributio of the Mea Review: The populatio is the etire collectio of all idividuals or objects of iterest The sample is the portio of the populatio

More information

Non-life insurance mathematics. Nils F. Haavardsson, University of Oslo and DNB Skadeforsikring

Non-life insurance mathematics. Nils F. Haavardsson, University of Oslo and DNB Skadeforsikring No-life isurace mathematics Nils F. Haavardsso, Uiversity of Oslo ad DNB Skadeforsikrig Mai issues so far Why does isurace work? How is risk premium defied ad why is it importat? How ca claim frequecy

More information

Chapter 5 O A Cojecture Of Erdíos Proceedigs NCUR VIII è1994è, Vol II, pp 794í798 Jeærey F Gold Departmet of Mathematics, Departmet of Physics Uiversity of Utah Do H Tucker Departmet of Mathematics Uiversity

More information

A Note on Sums of Greatest (Least) Prime Factors

A Note on Sums of Greatest (Least) Prime Factors It. J. Cotemp. Math. Scieces, Vol. 8, 203, o. 9, 423-432 HIKARI Ltd, www.m-hikari.com A Note o Sums of Greatest (Least Prime Factors Rafael Jakimczuk Divisio Matemática, Uiversidad Nacioal de Luá Bueos

More information

A RANDOM PERMUTATION MODEL ARISING IN CHEMISTRY

A RANDOM PERMUTATION MODEL ARISING IN CHEMISTRY J. Appl. Prob. 45, 060 070 2008 Prited i Eglad Applied Probability Trust 2008 A RANDOM PERMUTATION MODEL ARISING IN CHEMISTRY MARK BROWN, The City College of New York EROL A. PEKÖZ, Bosto Uiversity SHELDON

More information

The Stable Marriage Problem

The Stable Marriage Problem The Stable Marriage Problem William Hut Lae Departmet of Computer Sciece ad Electrical Egieerig, West Virgiia Uiversity, Morgatow, WV William.Hut@mail.wvu.edu 1 Itroductio Imagie you are a matchmaker,

More information

Lecture 3. denote the orthogonal complement of S k. Then. 1 x S k. n. 2 x T Ax = ( ) λ x. with x = 1, we have. i = λ k x 2 = λ k.

Lecture 3. denote the orthogonal complement of S k. Then. 1 x S k. n. 2 x T Ax = ( ) λ x. with x = 1, we have. i = λ k x 2 = λ k. 18.409 A Algorithmist s Toolkit September 17, 009 Lecture 3 Lecturer: Joatha Keler Scribe: Adre Wibisoo 1 Outlie Today s lecture covers three mai parts: Courat-Fischer formula ad Rayleigh quotiets The

More information

Perfect Packing Theorems and the Average-Case Behavior of Optimal and Online Bin Packing

Perfect Packing Theorems and the Average-Case Behavior of Optimal and Online Bin Packing SIAM REVIEW Vol. 44, No. 1, pp. 95 108 c 2002 Society for Idustrial ad Applied Mathematics Perfect Packig Theorems ad the Average-Case Behavior of Optimal ad Olie Bi Packig E. G. Coffma, Jr. C. Courcoubetis

More information

Overview of some probability distributions.

Overview of some probability distributions. Lecture Overview of some probability distributios. I this lecture we will review several commo distributios that will be used ofte throughtout the class. Each distributio is usually described by its probability

More information

5: Introduction to Estimation

5: Introduction to Estimation 5: Itroductio to Estimatio Cotets Acroyms ad symbols... 1 Statistical iferece... Estimatig µ with cofidece... 3 Samplig distributio of the mea... 3 Cofidece Iterval for μ whe σ is kow before had... 4 Sample

More information

PRODUCT RULE WINS A COMPETITIVE GAME

PRODUCT RULE WINS A COMPETITIVE GAME PROCEEDIGS OF THE AMERICA MATHEMATICAL SOCIETY Volume 00, umber 0, Pages 000 000 S 0002-9939XX0000-0 PRODUCT RULE WIS A COMPETITIVE GAME ADREW BEVERIDGE, TOM BOHMA, ALA FRIEZE, AD OLEG PIKHURKO Abstract

More information

Chapter 7 - Sampling Distributions. 1 Introduction. What is statistics? It consist of three major areas:

Chapter 7 - Sampling Distributions. 1 Introduction. What is statistics? It consist of three major areas: Chapter 7 - Samplig Distributios 1 Itroductio What is statistics? It cosist of three major areas: Data Collectio: samplig plas ad experimetal desigs Descriptive Statistics: umerical ad graphical summaries

More information

Ekkehart Schlicht: Economic Surplus and Derived Demand

Ekkehart Schlicht: Economic Surplus and Derived Demand Ekkehart Schlicht: Ecoomic Surplus ad Derived Demad Muich Discussio Paper No. 2006-17 Departmet of Ecoomics Uiversity of Muich Volkswirtschaftliche Fakultät Ludwig-Maximilias-Uiversität Müche Olie at http://epub.ub.ui-mueche.de/940/

More information

Running Time ( 3.1) Analysis of Algorithms. Experimental Studies ( 3.1.1) Limitations of Experiments. Pseudocode ( 3.1.2) Theoretical Analysis

Running Time ( 3.1) Analysis of Algorithms. Experimental Studies ( 3.1.1) Limitations of Experiments. Pseudocode ( 3.1.2) Theoretical Analysis Ruig Time ( 3.) Aalysis of Algorithms Iput Algorithm Output A algorithm is a step-by-step procedure for solvig a problem i a fiite amout of time. Most algorithms trasform iput objects ito output objects.

More information

1 Correlation and Regression Analysis

1 Correlation and Regression Analysis 1 Correlatio ad Regressio Aalysis I this sectio we will be ivestigatig the relatioship betwee two cotiuous variable, such as height ad weight, the cocetratio of a ijected drug ad heart rate, or the cosumptio

More information