Berry-Esseen bounds for self-normalized martingales

Size: px
Start display at page:

Download "Berry-Esseen bounds for self-normalized martingales"

Transcription

1 Berry-Essee bouds for self-ormalized martigales Xiequa Fa a, Qi-Ma Shao b a Ceter for Applied Mathematics, Tiaji Uiversity, Tiaji 30007, Chia b Departmet of Statistics, The Chiese Uiversity of Hog Kog, Shati, NT, Hog Kog Abstract A Berry-Essee boud is obtaied for self-ormalized martigales uder the assumptio of fiite momets. The boud coicides with the classical Berry-Essee boud for stadardized martigales. A example is give to show the optimality of the boud. Applicatios to Studet s statistic ad autoregressive process are also discussed. Keywords: Self-ormalized process, Berry-Essee bouds, martigales, Studet s statistic, autoregressive process 000 MSC: Primary 60G4; 60F05; Secodary 60E15 1. Itroductio Let X i i 1 be a sequece of idepedet o-degeerate real-valued radom variables with zero meas, ad let S = X i ad V = Xi be the partial sum ad the partial quadratic sum, respectively. The self-ormalized sum is defied as S /V. The study of the asymptotic behavior of self-ormalized sums has a log history. Whe X i i 1 are i.i.d. i the domai of ormal ad stable law, Loga et al. [11] obtaied the weak covergece for the self-ormalized sum, while Gié et al. [5] proved that S /V is asymptotically ormal if ad oly if X 1 belogs to the domai of attractio of a ormal law. Uder the same ecessary ad sufficiet coditio, Csörgő et al. [3] proved a self-ormalized type Dosker s theorem. For geeral idepedet radom variables with fiite + δ th momets, where 0 < δ 1, Betkus, Blozelis ad Götze [] see also Betkus ad Götze [1] for i.i.d. case have obtaied the followig Berry-Essee boud : If E X i +δ < for δ 0, 1], the these exists a absolute costat C such that sup x PS /V x Φx C B δ E X i +δ, Preprit submitted to Elsevier November 1, 017

2 where B = EX i, ad Φ x is the stadard ormal distributio fuctio. It is worth otig that the last boud coicides with the classical Berry-Essee boud for stadardized partial sums S /B ad it is the best possible. For the related error of PS /V x to 1 Φx, we refer to Shao [1], Jig, Shao ad Wag [9]. I these papers, self-ormalized Cramér type moderate deviatio theorems have bee established. We also refer to de la Peña, Lai ad Shao [4], Shao ad Wag [13] ad Shao ad Zhou [14] for surveys o recet developmets o self-ormalized limit theory. Despite the fact that the case for self-ormalized sums of idepedet radom variables is well studied, we are ot aware of Berry-Essee bouds for self-ormalized martigales i the literature. The mai purpose of this paper is to fill this gap. We first recall some Berry-Essee bouds for stadardized martigale differece sequece. Let X i, F i i=0,..., be a fiite sequece of martigale differeces defied o a probability space Ω, F, P, where X 0 = 0 ad {, Ω} = F 0... F F are icreasig σ-fields. Set S 0 = 0, S k = k X i, k = 1,...,. 1 The S = S k, F k k=0,..., is a martigale. Let [S] ad S be, respectively, the squared variace ad the coditioal variace of the martigale S, that is ad S 0 = 0, S k = [S] 0 = 0, [S] k = k X i k E[Xi F i 1 ], k = 1,...,. Suppose that E X i p < for some p > 1 ad all i = 1,...,. Defie N = E X i p + E S 1 p. Whe p 1, ], Heyde ad Brow [8] see also Theorem 3.10 of Hall ad Heyde [7] proved that there exits a costat C p depedig oly o p such that sup PS x Φ x C p N 1/p+1. 3 Later, Haeusler [6] gave a extesio of 3 to all p 1,. Moreover, Haeusler also gave a example to justify that his boud is asymptotically the best possible. It is remarked that the X i 1 i is stadardized, that is, EX i is close to 1.

3 I this paper, we prove that the Berry-Essee boud 3 also holds for self-ormalized martigales S / [S] ad ormalized martigales S / S. Moreover, we also justify the optimality of our bouds. Applicatios to Studet s statistic ad autoregressive process are discussed. The paper is orgaized as follows. Our mai results are stated ad discussed i Sectio. The applicatios are give i Sectio 3. Proofs of theorems are deferred to Sectio 4.. Mai results The followig theorem gives a couterpart of Haeusler s result [6] for self-ormalized martigales. Theorem.1. Suppose that E X i p < for some p > 1 ad all i = 1,...,. The there exits a costat C p depedig oly o p such that S sup P x Φ x C p N 1/p+1, 4 [S] where N is defied by. Moreover, there exit a sequece of martigale differeces X i, F i i=0,..., ad a positive costat c p depedig oly o p such that S sup P x Φ x N 1/p+1 c p. 5 [S] Clearly, iequality 5 shows that the boud 4 is asymptotically the best possible. For a statioary martigale differece sequece, the term E X i p is of order 1 p. The iequality 4 implies the followig corollary. Corollary.1. Let X i, F i i 1 be a statioary martigale differece sequece. Suppose that E X 1 p < for some p > 1. The there exits a costat c p, which does ot deped o, such that S sup P x Φ x c p 1 p + E S 1 p 1/p+1. 6 [S] The ext theorem gives a Berry-Essee boud for ormalized martigales S / S. Theorem.. Uder the assumptios of Theorem.1, the iequalities 4 ad 5 hold whe S / [S] is replaced by S / S. For a statioary martigale differece sequece, the followig result is a cosequece of the last theorem. Corollary.. Assume the coditios of Corollary.1. Iequality 6 holds whe S / [S] is replaced by S / S. 3

4 3. Applicatios 3.1. Applicatio to Studet s t-statistic The study of self-ormalized partial sums origiates from Studet s t-statistic. The Studet s t-statistic T is defied by T = X / σ, where X = S It is kow that for all x 0, P T > x ad σ = X i X. 1 S = P > x [S] + x 1 1/. Whe X i i 1 is a sequece of i.i.d. radom variables, Betkus ad Götze [1] proved that if E X i +δ < for all i = 1,..., ad some δ 0, 1], the sup PT x Φx = O δ/. 7 For martigales, we have the followig aalogue. Corollary 3.1. Let X i, F i i 1 be a statioary martigale differece sequece. Suppose that E X 1 p < for some p > 1. The there exits a costat C p, which does ot deped o, such that 6 holds whe PS / [S] x is replaced by PT x. 3.. Applicatio to autoregressive process Cosider the autoregressive process give by Y +1 = θy + ε +1, 0, where Y ad ε represet the observatio ad the drive oise, respectively. The parameter θ is ukow ad eeds to be estimated at stage from the data Y i, i. For sake of simplicity, we assume that Y 0 = 0. We also assume that ε 0 is a statioary martigale differece sequece with E[ε i ε 1,..., ε i 1 ] = σ a.s. for a positive costat σ. We ca estimate the ukow parameter θ by the least-squares estimator give by θ = Y iy i+1 Y. i It is well kow that θ θ Σ Y i coverges i distributio to a ormal law, see Theorem 3 of Lai ad Wei [10]. By Theorem., we have the followig Berry-Essee boud for the least-squares estimator θ. 4

5 Theorem 3.1. Suppose that E ε 1 p < for some p > 1. If θ < 1, the sup P θ θ Σ Y i xσ Φ x = O 1 p + p E Yi EYi p 1/p+1, 8 where Y = θ i ε i. 4. Proofs of theorems 4.1. Proof of Theorem.1 We assume that N 1. Otherwise, 4 is trivial. Firstly, we give a lower boud for PS x [S] Φx, x 0. Let ε 0, 1/] be a positive umber, whose exact value will be chose later. It is easy to see that for x 0, where PS x [S] Φx PS x [S], [S] < 1 + ε Φx PS x 1 + ε, [S] < 1 + ε Φx PS x 1 + ε P[S] 1 + ε Φx = I 1 + I I 3, 9 I 1 = PS x 1 + ε Φx 1 + ε, I = Φx 1 + ε Φx, I 3 = P[S] 1 + ε. Next, we estimate I 1, I ad I 3. By Haeusler s iequality [6] see also 3 whe p 1, ], we get the followig estimatio for I 1 : I 1 C p,1 E X i p + E S 1 p 1/p By oe-term Taylor s expasio, we have the followig estimatio for I : I c 1 e x / x 1 + ε 1 c ε. 11 5

6 For I 3, by Markov s iequality, it follows that I 3 = P[S] S + S 1 ε P [S] S ε + P S 1 ε c 3 ε p E [S] S p + E S 1 p. 1 We distiguish two cases to estimate I 3. Notice that [S] i S i, F i i=0,..., is also a martigale. Case 1 : If p 1, ], by the iequality of vo Bahr-Essee [15], it follows that Returig to 1, we have E [S] S p c 4 E Xi E[Xi F i 1 ] p I 3 c 6 ε p c 4 E[ X i p + E[Xi F i 1 ] p ] c 5 E X i p. 13 E X i p + E S 1. p 14 Case : If p >, by Rosethal s iequality cf. Theorem.1 of Hall ad Heyde [7], we have p/ E [S] S p C p, E E[Xi 4 F i 1 ] + E X i. p 15 Notig that X 4 i = X i p /p 1 X i p 1/p 1 for p >, we have by Hölder s iequality E[X 4 i F i 1 ] 1/p 1 p /p 1, E[ X i p F i 1 ] E[Xi F i 1 ] ad hece E[Xi 4 F i 1 ] 1/p 1 p /p 1 E[ X i p F i 1 ] E[Xi F i 1 ] 1/p 1 p /p 1. E[ X i p F i 1 ] S 6

7 By the iequality a + b q q a q + b q, a, b 0 ad q > 0, ad the fact that p p/p p, it follows that for p >, p/ E[Xi 4 F i 1 ] p p p/p p E[ X i p F i 1 ] S p/p p/p E[ X i p F i 1 ] 1 + S 1 p p/p p/p E[ X i p F i 1 ] p/p S + p E[ X i p F i 1 ] 1 pp /p. As to the secod term o the r.h.s. of the last iequality, we use the iequality ad hece Thus, x a y 1 a x + y, x, y 0 ad a [0, 1], p/ p/p E[Xi 4 F i 1 ] p E[ X i p F i 1 ] p/ E E[Xi 4 F i 1 ] + p E[ X i p F i 1 ] + S 1. p p/p ] [E p E[ X i p F i 1 ] + E X i p + E S 1 p 16 [ p/p p E X i p + E X i p + E S 1 ]. p 17 Returig to 15, we get for p >, p/p E [S] S p C p,3 E X i p + E S 1. p 7

8 From 1 ad the last iequality, we obtai for p >, p/p I 3 C p,4 ε p E X i p + E S 1. p 18 By the iequalities 14 ad 18, we always have for p > 1, I 3 C p,5 ε p E X i p + p/p E X i p + E S 1. p 19 Combiig 9, 10, 11 ad 19 together, we deduce that for p > 1, Takig PS x [S] Φx C p,1 E X i p + E S 1 p 1/p+1 c ε C p,5 ε p p/p E X i p + E X i p + E S 1. p p/p 1/p+1 ε = E X i p + E X i p + E S 1 p, 0 we obtai for x 0 ad p > 1, PS x [S] Φx C p,1 E X i p + E S 1 p 1/p+1 p/p 1/p+1 C p,6 E X i p + E X i p + E S 1 p C p,7 E X i p + E S 1 p 1/p+1, 1 where the last lie follows from the fact that p/p p + 1 1/p + 1 ad N 1. Secodly, we give a upper boud for PS x [S] Φx, x 0. It is obvious that for x 0, PS x [S] Φx 8

9 PS x [S], [S] > 1 ε Φx + PS x [S], [S] 1 ε PS x 1 ε, [S] > 1 ε Φx + P[S] 1 ε PS x 1 ε Φx 1 ε + Φx 1 ε Φx + P[S] 1 ε = I 4 + I 5 + I 6. Followig the same lies as i the proof of 1, we get for x 0 ad p > 1, PS x [S] Φx C p,8 E X i p + E S 1 p 1/p+1. Combiig 1 ad together, we get for p > 1, sup x 0 PS x [S] Φx C p,8 E X i p + E S 1 p 1/p+1. 3 Notice that S k, F k k=0,..., is also a martigale. Applyig the last iequality to S k, F k k=0,...,, we get sup PS x [S] Φx x>0 = sup PS x [S] Φx x>0 = sup Φ x P S < x [S] x>0 C p,9 E X i p + E S 1 p 1/p+1. 4 Combiig the iequalities 3 ad 4 together, we obtai sup PS x [S] Φx C p,10 E X i p + E S 1 p 1/p+1, 5 which gives the desired iequality 4. Next we give a proof of 5. We follow the example of Haeusler [6]. Let α 1 be a sequece of positive umbers such that α 0 as. Defie fuctio f : R [0, as follows { x f x = 1, if 1 α x <, 0, otherwise. 9

10 Furthermore, let X 1,..., X 1 be idepedet ad ormally distributed radom variables with mea 0 ad variace 1/ 1. Deote ν x the oe-poit mass cocetratio at x. Defie the radom variable X such that its coditioal distributio, give X 1,..., X 1, is PX S 1 = x = 1 ν α f x + 1 ν α f x, where S 1 = 1 X i. Deote F i the atural filtratio of X 1,..., X, that is F 0 beig the trivial σ field ad F i = σ{x 1,..., X i }, i = 1,...,. Clearly, X i, F i i=0,..., is a fiite sequece of martigale differeces. Moreover, it holds ad E X p = 1 E X i p = 1 1 E N 0, p 1 p C p,11 1 p = 1 π 1 y p PX dy N 0, 1 = xpn 0, 1 dx α α x p e 1 x dx C p,1 α p+ 1 6 for some costats 0 < C p,11, C p,1 <, where N 0, 1 is a stadard radom variable. Similarly, we have 1 1 E[Xi F i 1 ] = EXi = 1 ad E S 1 p = E E[X F 1 ] p = y PX dy N 0, 1 = x p PN 0, 1 dx = 1 π 1 α α x p e 1 x dx C p,1 α p Thus N C p α p+ 1, 10

11 where C p is a positive costat depedig oly o p. O the other had, we have S P 0 = P X + S 1 0 [S] Hece, we deduce that sup P P = = = 1 α + PX x N 0, 1 = xpn 0, 1 dx PX x N 0, 1 = xpn 0, 1 dx PX 1 x N 0, 1 = xpn 0, 1 dx α e 1 x dx + 1 α x dx π 1 0 π = Φ π S α 1 + o1 x Φ x N 1/p+1 [S] S 1 1 α. 1 e 0 Φ 0 C p α p /p+1 + o1 [S] = 1 4 α C p α p /p+1 + o1 π 1 4 π C p 1/p+1. This completes the proof of Theorem Proof of Theorem. First, we give a lower boud for PS x S Φx, x 0. Let ε 0, 1/] be a positive umber, whose exact value will be chose later. It is easy to see that for x 0, where PS x S Φx PS x S, S < 1 + ε Φx PS x 1 + ε, S < 1 + ε Φx PS x 1 + ε P S 1 + ε Φx = I 1 + I I 7, I 7 = P S 1 + ε. 11

12 For I 7, by Markov s iequality, it follows that I 7 ε p E S 1 p. 8 Combiig the estimatios 10, 11 ad 8 together, we have for p > 1, PS x [S] Φx C p,1 E X i p + E S 1 p 1/p+1 c ε ε p E S 1 p. Next we carry out a argumet as the proof of Theorem.1 with what we obtai is sup P S ε = E S 1 p 1/p+1, 9 x Φ x C p, E X i p + E S 1 p 1/p+1, 30 S that is iequality 4 holds whe S / [S] is replaced by S / S. The proof of optimality is similar to the proof of 5. This completes the proof of Theorem Proof of Theorem 3.1 The proof of theorem is based o Theorem.. It is easy to see that 1 σ θ θ Σ Y i = Y iε i+1 σ Y. i Notice that Y = θ i ε i. Set X i = Y i ε i+1 σ EY i ad F i = σ{ε k, 1 k i + 1}. The it is easy to see that X i, F i i=0,..., is a sequece of martigale differeces, ad that 1 σ θ θ Σ Y i = S. S Moreover, we have EY = θ i σ = 1 θ 1 θ σ 1

13 ad By Rosethal s iequality, we also have ad EY i = E Y p C p EY p + E Y i p C p 1 θ i 1 θ σ. E θ i ε i p 1 θ pσ C p p + 1 θ p 1 θ 1 θ E ε 1 p p 1 θ i pσ p + 1 θ p i 1 θ 1 θ E ε 1 p. p Thus E X i p EX i p C p If θ < 1, iequality 31 implies that E X i p It is obvious that EX i p C p 1 θ i pσ p + 1 θ p i / 1 θ 1 θ E ε 1 p p 1 C p 1 θ p + σ p 1 θ p + E ε 1 p / 1 θ p 1 E ε 1 p 1 θ p σ p i 1 p. 1 θ i 1 θ σ p. 31 σ p S = Y i EY. 3 By Theorem., we obtai sup P θ θ Σ Y i xσ Φ x C E X i p p,θ + E Y i EX i p EY 1 i = O 1 p + p E Yi EYi p 1/p+1. This completes the proof of theorem. 13 p 1/p+1

14 Ackowledgemets The research is partially supported by Hog Kog RGC GRF Fa has bee partially supported by the Natioal Natural Sciece Foudatio of Chia Grat os ad Referees [1] Betkus, V., Götze, F The Berry-Essee boud for Studet s statistic. A. Probab. 41: [] Betkus, V., Blozelis, M., Götze, F A Berry-Essée boud for Studet s statistic i the o-iid case. J. Theor. Probab. 93: [3] Csörgő, M., Szyszkowicz, B., Wag, Q Dosker s theorem for self-ormalized partial sums processes. A. Probab [4] de la Peña, V.H., Lai, T.L., Shao, Q.M Self-ormalized processes: limit theory ad statistical applicatios. Spriger, New York. [5] Gié, E., Götze, F., Maso, D.M Whe is the Studet t-statistic asymptotically stadard ormal? A. Probab [6] Haeusler, E O the rate of covergece i the cetral limit theorem for martigales with discrete ad cotiuous time. A. Probab. 161, [7] Hall, P., Heyde, C.C Martigale Limit Theory ad its Applicatios. Academic, New York. [8] Heyde, C.C., Brow, B.M O the depature from ormality of a certai class of martigales. A. Math. Statist. 41, [9] Jig, B.Y., Shao, Q.M., Wag, Q Self-ormalized Cramér-type large deviatios for idepedet radom variables. A. Probab. 314: [10] Lai, T.L., Wei, C.Z Least squares estimates i stochastic regressio models with applicatios to idetificatio ad cotrol of dyamic systems. A. Statist. 101: [11] Loga, B.F., Mallows, C.L., Rice, S.O., Shepp, L.A Limit distributios of self-ormalized sums. A. Probab. 1: [1] Shao, Q.M A Cramér type large deviatio result for Studet s t statistic. J. Theor. Probab. 1: [13] Shao, Q.M., Wag, Q.M Self-ormalized limit theorems: A survey. Probab. Survey 10, [14] Shao, Q.M., Zhou, W.X Self-ormalizatio: Tamig a wild populatio i a heavy-tailed world. Appl. Math. J. Chiese Uiv. 3, [15] vo Bahr, B., Essee, C.G Iequalities for the rth absolute momet of a sum of radom variables, 1 r. A. Math. Statist. 361:

Self-normalized deviation inequalities with application to t-statistic

Self-normalized deviation inequalities with application to t-statistic Self-ormalized deviatio iequalities with applicatio to t-statistic Xiequa Fa Ceter for Applied Mathematics, Tiaji Uiversity, 30007 Tiaji, Chia Abstract Let ξ i i 1 be a sequece of idepedet ad symmetric

More information

Asymptotic distribution of products of sums of independent random variables

Asymptotic distribution of products of sums of independent random variables Proc. Idia Acad. Sci. Math. Sci. Vol. 3, No., May 03, pp. 83 9. c Idia Academy of Scieces Asymptotic distributio of products of sums of idepedet radom variables YANLING WANG, SUXIA YAO ad HONGXIA DU ollege

More information

Central limit theorem and almost sure central limit theorem for the product of some partial sums

Central limit theorem and almost sure central limit theorem for the product of some partial sums Proc. Idia Acad. Sci. Math. Sci. Vol. 8, No. 2, May 2008, pp. 289 294. Prited i Idia Cetral it theorem ad almost sure cetral it theorem for the product of some partial sums YU MIAO College of Mathematics

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

Regression with an Evaporating Logarithmic Trend

Regression with an Evaporating Logarithmic Trend Regressio with a Evaporatig Logarithmic Tred Peter C. B. Phillips Cowles Foudatio, Yale Uiversity, Uiversity of Aucklad & Uiversity of York ad Yixiao Su Departmet of Ecoomics Yale Uiversity October 5,

More information

A note on self-normalized Dickey-Fuller test for unit root in autoregressive time series with GARCH errors

A note on self-normalized Dickey-Fuller test for unit root in autoregressive time series with GARCH errors Appl. Math. J. Chiese Uiv. 008, 3(): 97-0 A ote o self-ormalized Dickey-Fuller test for uit root i autoregressive time series with GARCH errors YANG Xiao-rog ZHANG Li-xi Abstract. I this article, the uit

More information

Self-normalized Cramér type Moderate Deviations for the Maximum of Sums

Self-normalized Cramér type Moderate Deviations for the Maximum of Sums Self-ormalized Cramér type Moderate Deviatios for the Maximum of Sums Weidog Liu 1, Qi-Ma Shao ad Qiyig Wag 3 Abstract Let X 1, X,... be idepedet radom variables with zero meas ad fiite variaces, ad let

More information

An almost sure invariance principle for trimmed sums of random vectors

An almost sure invariance principle for trimmed sums of random vectors Proc. Idia Acad. Sci. Math. Sci. Vol. 20, No. 5, November 200, pp. 6 68. Idia Academy of Scieces A almost sure ivariace priciple for trimmed sums of radom vectors KE-ANG FU School of Statistics ad Mathematics,

More information

Limit distributions for products of sums

Limit distributions for products of sums Statistics & Probability Letters 62 (23) 93 Limit distributios for products of sums Yogcheg Qi Departmet of Mathematics ad Statistics, Uiversity of Miesota-Duluth, Campus Ceter 4, 7 Uiversity Drive, Duluth,

More information

A Note on the Kolmogorov-Feller Weak Law of Large Numbers

A Note on the Kolmogorov-Feller Weak Law of Large Numbers Joural of Mathematical Research with Applicatios Mar., 015, Vol. 35, No., pp. 3 8 DOI:10.3770/j.iss:095-651.015.0.013 Http://jmre.dlut.edu.c A Note o the Kolmogorov-Feller Weak Law of Large Numbers Yachu

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

o f P r o b a b i l i t y Vol. 10 (2005), Paper no. 38, pages Journal URL ejpecp/

o f P r o b a b i l i t y Vol. 10 (2005), Paper no. 38, pages Journal URL   ejpecp/ E l e c t r o i c J o u r a l o f P r o b a b i l i t y Vol. 10 005, Paper o. 38, pages 160-185. Joural URL http://www.math.washigto.edu/ ejpecp/ Limit Theorems for Self-ormalized Large Deviatio 1 Qiyig

More information

Approximation theorems for localized szász Mirakjan operators

Approximation theorems for localized szász Mirakjan operators Joural of Approximatio Theory 152 (2008) 125 134 www.elsevier.com/locate/jat Approximatio theorems for localized szász Miraja operators Lise Xie a,,1, Tigfa Xie b a Departmet of Mathematics, Lishui Uiversity,

More information

Notes 19 : Martingale CLT

Notes 19 : Martingale CLT Notes 9 : Martigale CLT Math 733-734: Theory of Probability Lecturer: Sebastie Roch Refereces: [Bil95, Chapter 35], [Roc, Chapter 3]. Sice we have ot ecoutered weak covergece i some time, we first recall

More information

MAXIMAL INEQUALITIES AND STRONG LAW OF LARGE NUMBERS FOR AANA SEQUENCES

MAXIMAL INEQUALITIES AND STRONG LAW OF LARGE NUMBERS FOR AANA SEQUENCES Commu Korea Math Soc 26 20, No, pp 5 6 DOI 0434/CKMS20265 MAXIMAL INEQUALITIES AND STRONG LAW OF LARGE NUMBERS FOR AANA SEQUENCES Wag Xueju, Hu Shuhe, Li Xiaoqi, ad Yag Wezhi Abstract Let {X, } be a sequece

More information

ECONOMETRIC THEORY. MODULE XIII Lecture - 34 Asymptotic Theory and Stochastic Regressors

ECONOMETRIC THEORY. MODULE XIII Lecture - 34 Asymptotic Theory and Stochastic Regressors ECONOMETRIC THEORY MODULE XIII Lecture - 34 Asymptotic Theory ad Stochastic Regressors Dr. Shalabh Departmet of Mathematics ad Statistics Idia Istitute of Techology Kapur Asymptotic theory The asymptotic

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

h(x i,x j ). (1.1) 1 i<j n

h(x i,x j ). (1.1) 1 i<j n ESAIM: PS 15 2011 168 179 DOI: 10.1051/ps/2009014 ESAIM: Probability ad Statistics www.esaim-ps.org CRAMÉR TYPE MODERATE DEVIATIONS FOR STUDENTIZED U-STATISTICS,, Tze Leg Lai 1, Qi-Ma Shao 2 ad Qiyig Wag

More information

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 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

2.2. Central limit theorem.

2.2. Central limit theorem. 36.. Cetral limit theorem. The most ideal case of the CLT is that the radom variables are iid with fiite variace. Although it is a special case of the more geeral Lideberg-Feller CLT, it is most stadard

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

LECTURE 8: ASYMPTOTICS I

LECTURE 8: ASYMPTOTICS I LECTURE 8: ASYMPTOTICS I We are iterested i the properties of estimators as. Cosider a sequece of radom variables {, X 1}. N. M. Kiefer, Corell Uiversity, Ecoomics 60 1 Defiitio: (Weak covergece) A sequece

More information

Entropy Rates and Asymptotic Equipartition

Entropy Rates and Asymptotic Equipartition Chapter 29 Etropy Rates ad Asymptotic Equipartitio Sectio 29. itroduces the etropy rate the asymptotic etropy per time-step of a stochastic process ad shows that it is well-defied; ad similarly for iformatio,

More information

Research Article Moment Inequality for ϕ-mixing Sequences and Its Applications

Research Article Moment Inequality for ϕ-mixing Sequences and Its Applications Hidawi Publishig Corporatio Joural of Iequalities ad Applicatios Volume 2009, Article ID 379743, 2 pages doi:0.55/2009/379743 Research Article Momet Iequality for ϕ-mixig Sequeces ad Its Applicatios Wag

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

Mi-Hwa Ko and Tae-Sung Kim

Mi-Hwa Ko and Tae-Sung Kim J. Korea Math. Soc. 42 2005), No. 5, pp. 949 957 ALMOST SURE CONVERGENCE FOR WEIGHTED SUMS OF NEGATIVELY ORTHANT DEPENDENT RANDOM VARIABLES Mi-Hwa Ko ad Tae-Sug Kim Abstract. For weighted sum of a sequece

More information

EFFECTIVE WLLN, SLLN, AND CLT IN STATISTICAL MODELS

EFFECTIVE WLLN, SLLN, AND CLT IN STATISTICAL MODELS EFFECTIVE WLLN, SLLN, AND CLT IN STATISTICAL MODELS Ryszard Zieliński Ist Math Polish Acad Sc POBox 21, 00-956 Warszawa 10, Polad e-mail: rziel@impagovpl ABSTRACT Weak laws of large umbers (W LLN), strog

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

Kolmogorov-Smirnov type Tests for Local Gaussianity in High-Frequency Data

Kolmogorov-Smirnov type Tests for Local Gaussianity in High-Frequency Data Proceedigs 59th ISI World Statistics Cogress, 5-30 August 013, Hog Kog (Sessio STS046) p.09 Kolmogorov-Smirov type Tests for Local Gaussiaity i High-Frequecy Data George Tauche, Duke Uiversity Viktor Todorov,

More information

MATH 472 / SPRING 2013 ASSIGNMENT 2: DUE FEBRUARY 4 FINALIZED

MATH 472 / SPRING 2013 ASSIGNMENT 2: DUE FEBRUARY 4 FINALIZED MATH 47 / SPRING 013 ASSIGNMENT : DUE FEBRUARY 4 FINALIZED Please iclude a cover sheet that provides a complete setece aswer to each the followig three questios: (a) I your opiio, what were the mai ideas

More information

Asymptotic Results for the Linear Regression Model

Asymptotic Results for the Linear Regression Model Asymptotic Results for the Liear Regressio Model C. Fli November 29, 2000 1. Asymptotic Results uder Classical Assumptios The followig results apply to the liear regressio model y = Xβ + ε, where X is

More information

Introduction to Extreme Value Theory Laurens de Haan, ISM Japan, Erasmus University Rotterdam, NL University of Lisbon, PT

Introduction to Extreme Value Theory Laurens de Haan, ISM Japan, Erasmus University Rotterdam, NL University of Lisbon, PT Itroductio to Extreme Value Theory Laures de Haa, ISM Japa, 202 Itroductio to Extreme Value Theory Laures de Haa Erasmus Uiversity Rotterdam, NL Uiversity of Lisbo, PT Itroductio to Extreme Value Theory

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

First Year Quantitative Comp Exam Spring, Part I - 203A. f X (x) = 0 otherwise

First Year Quantitative Comp Exam Spring, Part I - 203A. f X (x) = 0 otherwise First Year Quatitative Comp Exam Sprig, 2012 Istructio: There are three parts. Aswer every questio i every part. Questio I-1 Part I - 203A A radom variable X is distributed with the margial desity: >

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

Let us give one more example of MLE. Example 3. The uniform distribution U[0, θ] on the interval [0, θ] has p.d.f.

Let us give one more example of MLE. Example 3. The uniform distribution U[0, θ] on the interval [0, θ] has p.d.f. Lecture 5 Let us give oe more example of MLE. Example 3. The uiform distributio U[0, ] o the iterval [0, ] has p.d.f. { 1 f(x =, 0 x, 0, otherwise The likelihood fuctio ϕ( = f(x i = 1 I(X 1,..., X [0,

More information

ON POINTWISE BINOMIAL APPROXIMATION

ON POINTWISE BINOMIAL APPROXIMATION Iteratioal Joural of Pure ad Applied Mathematics Volume 71 No. 1 2011, 57-66 ON POINTWISE BINOMIAL APPROXIMATION BY w-functions K. Teerapabolar 1, P. Wogkasem 2 Departmet of Mathematics Faculty of Sciece

More information

Lecture 2: Concentration Bounds

Lecture 2: Concentration Bounds CSE 52: Desig ad Aalysis of Algorithms I Sprig 206 Lecture 2: Cocetratio Bouds Lecturer: Shaya Oveis Ghara March 30th Scribe: Syuzaa Sargsya Disclaimer: These otes have ot bee subjected to the usual scrutiy

More information

STAT331. Example of Martingale CLT with Cox s Model

STAT331. Example of Martingale CLT with Cox s Model STAT33 Example of Martigale CLT with Cox s Model I this uit we illustrate the Martigale Cetral Limit Theorem by applyig it to the partial likelihood score fuctio from Cox s model. For simplicity of presetatio

More information

A Note on Sums of Independent Random Variables

A Note on Sums of Independent Random Variables Cotemorary Mathematics Volume 00 XXXX A Note o Sums of Ideedet Radom Variables Pawe l Hitczeko ad Stehe Motgomery-Smith Abstract I this ote a two sided boud o the tail robability of sums of ideedet ad

More information

Invariance principles for standard-normalized and self-normalized random fields

Invariance principles for standard-normalized and self-normalized random fields Ivariace priciples for stadard-ormalized ad self-ormalized radom fields M. El Machkouri, L. Ouchti 7th February 005 Abstract We ivestigate the ivariace priciple for set-idexed partial sums of a statioary

More information

Gamma Distribution and Gamma Approximation

Gamma Distribution and Gamma Approximation Gamma Distributio ad Gamma Approimatio Xiaomig Zeg a Fuhua (Frak Cheg b a Xiame Uiversity, Xiame 365, Chia mzeg@jigia.mu.edu.c b Uiversity of Ketucky, Leigto, Ketucky 456-46, USA cheg@cs.uky.edu Abstract

More information

SDS 321: Introduction to Probability and Statistics

SDS 321: Introduction to Probability and Statistics SDS 321: Itroductio to Probability ad Statistics Lecture 23: Cotiuous radom variables- Iequalities, CLT Puramrita Sarkar Departmet of Statistics ad Data Sciece The Uiversity of Texas at Austi www.cs.cmu.edu/

More information

Notes 5 : More on the a.s. convergence of sums

Notes 5 : More on the a.s. convergence of sums Notes 5 : More o the a.s. covergece of sums Math 733-734: Theory of Probability Lecturer: Sebastie Roch Refereces: Dur0, Sectios.5; Wil9, Sectio 4.7, Shi96, Sectio IV.4, Dur0, Sectio.. Radom series. Three-series

More information

Solution to Chapter 2 Analytical Exercises

Solution to Chapter 2 Analytical Exercises Nov. 25, 23, Revised Dec. 27, 23 Hayashi Ecoometrics Solutio to Chapter 2 Aalytical Exercises. For ay ε >, So, plim z =. O the other had, which meas that lim E(z =. 2. As show i the hit, Prob( z > ε =

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

Diagonal approximations by martingales

Diagonal approximations by martingales Alea 7, 257 276 200 Diagoal approximatios by martigales Jaa Klicarová ad Dalibor Volý Faculty of Ecoomics, Uiversity of South Bohemia, Studetsa 3, 370 05, Cese Budejovice, Czech Republic E-mail address:

More information

Generalized Law of the Iterated Logarithm and Its Convergence Rate

Generalized Law of the Iterated Logarithm and Its Convergence Rate Stochastic Aalysis ad Applicatios, 25: 89 03, 2007 Copyright Taylor & Fracis Group, LLC ISSN 0736-2994 prit/532-9356 olie DOI: 0.080/073629906005997 Geeralized Law of the Iterated Logarithm ad Its Covergece

More information

The log-behavior of n p(n) and n p(n)/n

The log-behavior of n p(n) and n p(n)/n Ramauja J. 44 017, 81-99 The log-behavior of p ad p/ William Y.C. Che 1 ad Ke Y. Zheg 1 Ceter for Applied Mathematics Tiaji Uiversity Tiaji 0007, P. R. Chia Ceter for Combiatorics, LPMC Nakai Uivercity

More information

Precise Rates in Complete Moment Convergence for Negatively Associated Sequences

Precise Rates in Complete Moment Convergence for Negatively Associated Sequences Commuicatios of the Korea Statistical Society 29, Vol. 16, No. 5, 841 849 Precise Rates i Complete Momet Covergece for Negatively Associated Sequeces Dae-Hee Ryu 1,a a Departmet of Computer Sciece, ChugWoo

More information

Expectation and Variance of a random variable

Expectation and Variance of a random variable Chapter 11 Expectatio ad Variace of a radom variable The aim of this lecture is to defie ad itroduce mathematical Expectatio ad variace of a fuctio of discrete & cotiuous radom variables ad the distributio

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

Discrete Mathematics for CS Spring 2008 David Wagner Note 22

Discrete Mathematics for CS Spring 2008 David Wagner Note 22 CS 70 Discrete Mathematics for CS Sprig 2008 David Wager Note 22 I.I.D. Radom Variables Estimatig the bias of a coi Questio: We wat to estimate the proportio p of Democrats i the US populatio, by takig

More information

RATES OF APPROXIMATION IN THE MULTIDIMENSIONAL INVARIANCE PRINCIPLE FOR SUMS OF I.I.D. RANDOM VECTORS WITH FINITE MOMENTS

RATES OF APPROXIMATION IN THE MULTIDIMENSIONAL INVARIANCE PRINCIPLE FOR SUMS OF I.I.D. RANDOM VECTORS WITH FINITE MOMENTS RATES OF APPROXIMATION IN THE MULTIDIMENSIONAL INVARIANCE PRINCIPLE FOR SUMS OF I.I.D. RANDOM VECTORS WITH FINITE MOMENTS F. Götze 1 ad A. Yu. Zaitsev 1,2 Uiversity of Bielefeld 1 St. Petersburg Departmet

More information

A UNIFIED APPROACH TO EDGEWORTH EXPANSIONS FOR A GENERAL CLASS OF STATISTICS

A UNIFIED APPROACH TO EDGEWORTH EXPANSIONS FOR A GENERAL CLASS OF STATISTICS Statistica Siica 20 2010, 613-636 A UNIFIED APPROACH TO EDGEWORTH EXPANSIONS FOR A GENERAL CLASS OF STATISTICS Big-Yi Jig ad Qiyig Wag Hog Kog Uiversity of Sciece ad Techology ad Uiversity of Sydey Abstract:

More information

Complete Convergence for Asymptotically Almost Negatively Associated Random Variables

Complete Convergence for Asymptotically Almost Negatively Associated Random Variables Applied Mathematical Scieces, Vol. 12, 2018, o. 30, 1441-1452 HIKARI Ltd, www.m-hikari.com https://doi.org/10.12988/ams.2018.810142 Complete Covergece for Asymptotically Almost Negatively Associated Radom

More information

32 estimating the cumulative distribution function

32 estimating the cumulative distribution function 32 estimatig the cumulative distributio fuctio 4.6 types of cofidece itervals/bads Let F be a class of distributio fuctios F ad let θ be some quatity of iterest, such as the mea of F or the whole fuctio

More information

TESTING FOR THE BUFFERED AUTOREGRESSIVE PROCESSES (SUPPLEMENTARY MATERIAL)

TESTING FOR THE BUFFERED AUTOREGRESSIVE PROCESSES (SUPPLEMENTARY MATERIAL) TESTING FOR THE BUFFERED AUTOREGRESSIVE PROCESSES SUPPLEMENTARY MATERIAL) By Ke Zhu, Philip L.H. Yu ad Wai Keug Li Chiese Academy of Scieces ad Uiversity of Hog Kog APPENDIX: PROOFS I this appedix, we

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

Mathematical Statistics - MS

Mathematical Statistics - MS Paper Specific Istructios. The examiatio is of hours duratio. There are a total of 60 questios carryig 00 marks. The etire paper is divided ito three sectios, A, B ad C. All sectios are compulsory. Questios

More information

Direction: This test is worth 250 points. You are required to complete this test within 50 minutes.

Direction: This test is worth 250 points. You are required to complete this test within 50 minutes. Term Test October 3, 003 Name Math 56 Studet Number Directio: This test is worth 50 poits. You are required to complete this test withi 50 miutes. I order to receive full credit, aswer each problem completely

More information

ECE 901 Lecture 12: Complexity Regularization and the Squared Loss

ECE 901 Lecture 12: Complexity Regularization and the Squared Loss ECE 90 Lecture : Complexity Regularizatio ad the Squared Loss R. Nowak 5/7/009 I the previous lectures we made use of the Cheroff/Hoeffdig bouds for our aalysis of classifier errors. Hoeffdig s iequality

More information

A RANK STATISTIC FOR NON-PARAMETRIC K-SAMPLE AND CHANGE POINT PROBLEMS

A RANK STATISTIC FOR NON-PARAMETRIC K-SAMPLE AND CHANGE POINT PROBLEMS J. Japa Statist. Soc. Vol. 41 No. 1 2011 67 73 A RANK STATISTIC FOR NON-PARAMETRIC K-SAMPLE AND CHANGE POINT PROBLEMS Yoichi Nishiyama* We cosider k-sample ad chage poit problems for idepedet data i a

More information

An alternative proof of a theorem of Aldous. concerning convergence in distribution for martingales.

An alternative proof of a theorem of Aldous. concerning convergence in distribution for martingales. A alterative proof of a theorem of Aldous cocerig covergece i distributio for martigales. Maurizio Pratelli Dipartimeto di Matematica, Uiversità di Pisa. Via Buoarroti 2. I-56127 Pisa, Italy e-mail: pratelli@dm.uipi.it

More information

EECS564 Estimation, Filtering, and Detection Hwk 2 Solns. Winter p θ (z) = (2θz + 1 θ), 0 z 1

EECS564 Estimation, Filtering, and Detection Hwk 2 Solns. Winter p θ (z) = (2θz + 1 θ), 0 z 1 EECS564 Estimatio, Filterig, ad Detectio Hwk 2 Sols. Witer 25 4. Let Z be a sigle observatio havig desity fuctio where. p (z) = (2z + ), z (a) Assumig that is a oradom parameter, fid ad plot the maximum

More information

The standard deviation of the mean

The standard deviation of the mean Physics 6C Fall 20 The stadard deviatio of the mea These otes provide some clarificatio o the distictio betwee the stadard deviatio ad the stadard deviatio of the mea.. The sample mea ad variace Cosider

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

Random Variables, Sampling and Estimation

Random Variables, Sampling and Estimation Chapter 1 Radom Variables, Samplig ad Estimatio 1.1 Itroductio This chapter will cover the most importat basic statistical theory you eed i order to uderstad the ecoometric material that will be comig

More information

Equivalent Conditions of Complete Convergence and Complete Moment Convergence for END Random Variables

Equivalent Conditions of Complete Convergence and Complete Moment Convergence for END Random Variables Chi. A. Math. Ser. B 391, 2018, 83 96 DOI: 10.1007/s11401-018-1053-9 Chiese Aals of Mathematics, Series B c The Editorial Office of CAM ad Spriger-Verlag Berli Heidelberg 2018 Equivalet Coditios of Complete

More information

Rates of Convergence by Moduli of Continuity

Rates of Convergence by Moduli of Continuity Rates of Covergece by Moduli of Cotiuity Joh Duchi: Notes for Statistics 300b March, 017 1 Itroductio I this ote, we give a presetatio showig the importace, ad relatioship betwee, the modulis of cotiuity

More information

Research Article On the Strong Laws for Weighted Sums of ρ -Mixing Random Variables

Research Article On the Strong Laws for Weighted Sums of ρ -Mixing Random Variables Hidawi Publishig Corporatio Joural of Iequalities ad Applicatios Volume 2011, Article ID 157816, 8 pages doi:10.1155/2011/157816 Research Article O the Strog Laws for Weighted Sums of ρ -Mixig Radom Variables

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

Linear regression. Daniel Hsu (COMS 4771) (y i x T i β)2 2πσ. 2 2σ 2. 1 n. (x T i β y i ) 2. 1 ˆβ arg min. β R n d

Linear regression. Daniel Hsu (COMS 4771) (y i x T i β)2 2πσ. 2 2σ 2. 1 n. (x T i β y i ) 2. 1 ˆβ arg min. β R n d Liear regressio Daiel Hsu (COMS 477) Maximum likelihood estimatio Oe of the simplest liear regressio models is the followig: (X, Y ),..., (X, Y ), (X, Y ) are iid radom pairs takig values i R d R, ad Y

More information

Probability and Statistics

Probability and Statistics ICME Refresher Course: robability ad Statistics Staford Uiversity robability ad Statistics Luyag Che September 20, 2016 1 Basic robability Theory 11 robability Spaces A probability space is a triple (Ω,

More information

A survey on penalized empirical risk minimization Sara A. van de Geer

A survey on penalized empirical risk minimization Sara A. van de Geer A survey o pealized empirical risk miimizatio Sara A. va de Geer We address the questio how to choose the pealty i empirical risk miimizatio. Roughly speakig, this pealty should be a good boud for the

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

5.1 A mutual information bound based on metric entropy

5.1 A mutual information bound based on metric entropy Chapter 5 Global Fao Method I this chapter, we exted the techiques of Chapter 2.4 o Fao s method the local Fao method) to a more global costructio. I particular, we show that, rather tha costructig a local

More information

MA Advanced Econometrics: Properties of Least Squares Estimators

MA Advanced Econometrics: Properties of Least Squares Estimators MA Advaced Ecoometrics: Properties of Least Squares Estimators Karl Whela School of Ecoomics, UCD February 5, 20 Karl Whela UCD Least Squares Estimators February 5, 20 / 5 Part I Least Squares: Some Fiite-Sample

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

A statistical method to determine sample size to estimate characteristic value of soil parameters

A statistical method to determine sample size to estimate characteristic value of soil parameters A statistical method to determie sample size to estimate characteristic value of soil parameters Y. Hojo, B. Setiawa 2 ad M. Suzuki 3 Abstract Sample size is a importat factor to be cosidered i determiig

More information

Lecture 4. We also define the set of possible values for the random walk as the set of all x R d such that P(S n = x) > 0 for some n.

Lecture 4. We also define the set of possible values for the random walk as the set of all x R d such that P(S n = x) > 0 for some n. Radom Walks ad Browia Motio Tel Aviv Uiversity Sprig 20 Lecture date: Mar 2, 20 Lecture 4 Istructor: Ro Peled Scribe: Lira Rotem This lecture deals primarily with recurrece for geeral radom walks. We preset

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

Supplementary Material for Fast Stochastic AUC Maximization with O(1/n)-Convergence Rate

Supplementary Material for Fast Stochastic AUC Maximization with O(1/n)-Convergence Rate Supplemetary Material for Fast Stochastic AUC Maximizatio with O/-Covergece Rate Migrui Liu Xiaoxua Zhag Zaiyi Che Xiaoyu Wag 3 iabao Yag echical Lemmas ized versio of Hoeffdig s iequality, ote that We

More information

On the convergence rates of Gladyshev s Hurst index estimator

On the convergence rates of Gladyshev s Hurst index estimator Noliear Aalysis: Modellig ad Cotrol, 2010, Vol 15, No 4, 445 450 O the covergece rates of Gladyshev s Hurst idex estimator K Kubilius 1, D Melichov 2 1 Istitute of Mathematics ad Iformatics, Vilius Uiversity

More information

Lecture 12: September 27

Lecture 12: September 27 36-705: Itermediate Statistics Fall 207 Lecturer: Siva Balakrisha Lecture 2: September 27 Today we will discuss sufficiecy i more detail ad the begi to discuss some geeral strategies for costructig estimators.

More information

Notes 27 : Brownian motion: path properties

Notes 27 : Brownian motion: path properties Notes 27 : Browia motio: path properties Math 733-734: Theory of Probability Lecturer: Sebastie Roch Refereces:[Dur10, Sectio 8.1], [MP10, Sectio 1.1, 1.2, 1.3]. Recall: DEF 27.1 (Covariace) Let X = (X

More information

Parameter, Statistic and Random Samples

Parameter, Statistic and Random Samples Parameter, Statistic ad Radom Samples A parameter is a umber that describes the populatio. It is a fixed umber, but i practice we do ot kow its value. A statistic is a fuctio of the sample data, i.e.,

More information

ON THE KOMLÓS RÉVÉSZ ESTIMATION PROBLEM FOR RANDOM VARIABLES WITHOUT VARIANCES. Ben-Gurion University, Israel

ON THE KOMLÓS RÉVÉSZ ESTIMATION PROBLEM FOR RANDOM VARIABLES WITHOUT VARIANCES. Ben-Gurion University, Israel ON THE KOMLÓS RÉVÉSZ ESTIMATION PROBLEM FOR RANDOM VARIABLES WITHOUT VARIANCES GUY COHEN Be-Gurio Uiversity, Israel e-mail: guycohe@ee.bgu.ac.il Abstract. Let {X } L p P), 1 < p 2, q = p/p 1), be a sequece

More information

Mixingales. Chapter 7

Mixingales. Chapter 7 Chapter 7 Mixigales I this sectio we prove some of the results stated i the previous sectios usig mixigales. We first defie a mixigale, otig that the defiitio we give is ot the most geeral defiitio. Defiitio

More information

A central limit theorem for moving average process with negatively associated innovation

A central limit theorem for moving average process with negatively associated innovation Iteratioal Mathematical Forum, 1, 2006, o. 10, 495-502 A cetral limit theorem for movig average process with egatively associated iovatio MI-HWA KO Statistical Research Ceter for Complex Systems Seoul

More information

Chapter 6 Principles of Data Reduction

Chapter 6 Principles of Data Reduction Chapter 6 for BST 695: Special Topics i Statistical Theory. Kui Zhag, 0 Chapter 6 Priciples of Data Reductio Sectio 6. Itroductio Goal: To summarize or reduce the data X, X,, X to get iformatio about a

More information

January 25, 2017 INTRODUCTION TO MATHEMATICAL STATISTICS

January 25, 2017 INTRODUCTION TO MATHEMATICAL STATISTICS Jauary 25, 207 INTRODUCTION TO MATHEMATICAL STATISTICS Abstract. A basic itroductio to statistics assumig kowledge of probability theory.. Probability I a typical udergraduate problem i probability, we

More information

Basis for simulation techniques

Basis for simulation techniques Basis for simulatio techiques M. Veeraraghava, March 7, 004 Estimatio is based o a collectio of experimetal outcomes, x, x,, x, where each experimetal outcome is a value of a radom variable. x i. Defiitios

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

Econ 325/327 Notes on Sample Mean, Sample Proportion, Central Limit Theorem, Chi-square Distribution, Student s t distribution 1.

Econ 325/327 Notes on Sample Mean, Sample Proportion, Central Limit Theorem, Chi-square Distribution, Student s t distribution 1. Eco 325/327 Notes o Sample Mea, Sample Proportio, Cetral Limit Theorem, Chi-square Distributio, Studet s t distributio 1 Sample Mea By Hiro Kasahara We cosider a radom sample from a populatio. Defiitio

More information

arxiv: v1 [math.pr] 13 Oct 2011

arxiv: v1 [math.pr] 13 Oct 2011 A tail iequality for quadratic forms of subgaussia radom vectors Daiel Hsu, Sham M. Kakade,, ad Tog Zhag 3 arxiv:0.84v math.pr] 3 Oct 0 Microsoft Research New Eglad Departmet of Statistics, Wharto School,

More information

Random Walks on Discrete and Continuous Circles. by Jeffrey S. Rosenthal School of Mathematics, University of Minnesota, Minneapolis, MN, U.S.A.

Random Walks on Discrete and Continuous Circles. by Jeffrey S. Rosenthal School of Mathematics, University of Minnesota, Minneapolis, MN, U.S.A. Radom Walks o Discrete ad Cotiuous Circles by Jeffrey S. Rosethal School of Mathematics, Uiversity of Miesota, Mieapolis, MN, U.S.A. 55455 (Appeared i Joural of Applied Probability 30 (1993), 780 789.)

More information

The variance of a sum of independent variables is the sum of their variances, since covariances are zero. Therefore. V (xi )= n n 2 σ2 = σ2.

The variance of a sum of independent variables is the sum of their variances, since covariances are zero. Therefore. V (xi )= n n 2 σ2 = σ2. SAMPLE STATISTICS A radom sample x 1,x,,x from a distributio f(x) is a set of idepedetly ad idetically variables with x i f(x) for all i Their joit pdf is f(x 1,x,,x )=f(x 1 )f(x ) f(x )= f(x i ) The sample

More information

A Hilbert Space Central Limit Theorem for Geometrically Ergodic Markov Chains

A Hilbert Space Central Limit Theorem for Geometrically Ergodic Markov Chains A Hilbert Space Cetral Limit Theorem for Geometrically Ergodic Marov Chais Joh Stachursi Research School of Ecoomics, Australia Natioal Uiversity Abstract This ote proves a simple but useful cetral limit

More information