Extreme value theory of mixture generalized order statistics

Size: px
Start display at page:

Download "Extreme value theory of mixture generalized order statistics"

Transcription

1 ProbStat Forum, Volume, July 208, Pages 04 6 ISSN ProbStat Forum is an e-journal. For details please visit.probstat.org.in Extreme value theory of mixture generalized order statistics A M Elsaah a, Gajendra K Vishaarma b, Zhongquan Tan c a Department of Mathematics, Faculty of Science, Zagazig University, Zagazig 4459, Egypt and Division of Science and Technology, BNU-HKBU United International College, Zhuhai 59085, China. b Department of Applied Mathematics, Indian Institute of Technology, Dhanbad , India. c College of Mathematics, Physics and Information Engineering, Jiaxing University, Jiaxing 3400, China. Abstract. Most statistical approaches assume that all of the data points come from the same distribution. Hoever, in real-life applications the data points come from more than one distribution ith no information to identify hich observation goes ith hich distribution. In such cases, the classical extreme value theory can not help us. In this paper, e investigate the extreme value theory of data from more than one distribution based on generalized order statistics under continuous strictly monotone normalization. This paper investigates the asymptotic behaviors of upper and loer extremes generalized order statistics based on a random sample dran from a finite mixture of distributions normalized by the same continuous strictly monotonic sequence or a mixture of continuous strictly monotonic sequences. For illustration of the usage of our theoretical results, these asymptotic behaviors under linear and poer normalization are studied ith examples.. Introduction.. Generalized order statistics Let X j : j N} be a sequence of independent and identically distributed random variables ith common probability density function f and distribution function F. If the first n random variables are arranged in ascending order of magnitude and ritten as X :n < X 2:n <... < X n:n, e call them ordinary order statistics. Generalized order statistics have been introduced by Kamps (995 as a unification of several models of ascendingly ordered random variables. The generalized order statistics X ( m, :n, X ( m, 2:n,..., X n:n ( m, are defined by their probability density function, hich is given on the cone (x,..., x n : x 0 = F (0 < x... x n < F ( = x 0 } as follos n f ( m,,2,...,n:n (x,..., x n = f(x n ( F(x n θ i f(x i ( F(x i mi, here θ,..., θ n are defined by θ n = > 0 and θ j m = (m,..., m n R n. = + n j + n i=j m i > 0, j =, 2,..., n and 200 Mathematics Subject Classification. 62G30, 60F05, 62E20. Keyords. Mixture distributions; Extreme value theory; Generalizes order statistics; Linear normalization; Poer normalization; Domains of attraction. Received: 3 December 207; Revised: 03 April 208, Re-revised 07 May 208; Accepted: 3 June 208. Elsaah-UIC Grants (Nos: R20409, R2072 and R2080 & the Zhuhai Premier Discipline Grant and Tan-National Science Foundation of China (No & Natural Science Foundation of Zhejiang Province of China (No. LQ4A address: a_elsaah85@yahoo.com, amelsaah@uic.edu.h, a.elsaah@zu.edu.eg (A M Elsaah

2 Elsaah, Vishaarma and Tan / ProbStat Forum, Volume, July 208, Pages Generalized order statistics have been idely used and they attracted much attention since Kamps (995 first introduced them. Generalized order statistics have the advantage of including important ell-non models hich are discussed separately in literature, such as the ordinary order statistics, sequential order statistics, Progressive type II censored order statistics, record values, th record values and Pfeifer s records. Particular choice of the parameters θ,..., θ n leads to different models, e.g., ordinary order statistics (m =... = m n = 0, = ; order statistics ith non-integral sample size (m =... = m n = 0, = τ n +, and τ > n ; th record values (m =... = m n = and is any positive integer and sequential order statistics (m i = (n i + φ i (n iφ i+, i n, = φ n and φ, φ 2,..., φ n > 0. In this paper, e consider a ide subclass of generalized order statistics by assuming θ j θ j+ = m+ > 0, i.e., m =... = m n = m. This subclass is non as m-generalized order statistics, e call it symmetric generalized order statistics throughout this paper. Many important practical models, such as ordinary order statistics, record values, sequential order statistics and order statistics ith non-integral sample size, are included in the symmetric generalized order statistics. The distribution function of the lth loer and the lth upper symmetric generalized order statistics are represented by and (x = I ( F(x m+ ( l, m + + n l n l+:n (x = I ( F(x (n m+ l +, m + + l, (2 x respectively, here I x (n, m = B(n,m 0 tn ( t m dt, B(n, m = (n(m (n+m function (c.f. Nasri-Roudsari, Classical extreme value theory ( and (. is the gamma The main problem of the classical extreme value theory of ordinary order statistics is to find conditions on the distribution function F under hich there is a strongly monotone continuous transformation B n (x, such that for the lth upper extreme ordinary order statistic the distribution function of Bn (X n l+:n (for the lth loer extreme ordinary order statistic the distribution function of Bn (X converges ealy to a non-degenerate distribution (i.e., the distribution is not degenerate at 0 or, to obtain the classes of possible limit distribution functions and to give methods for calculating the normalizing transformation B n (x. Several authors have studied this problem for the linear transformation B n (x = a n x + b n, such as Gnedeno (943, Balema and de Haan (972 and Smirnov (952. The central result of the classical extreme value theory turns out that the class of possible limit distributions has essentially three different types of distributions. For some suitable normalizing constants c n, a n > 0 and d n, b n R, e have Ξ (0, (c nx + d n H l(0, i,α (x = (l U i,α(x t l e t dt (3 and Ξ (0, n l+:n (a nx + b n H u(0, j,β (x = t l e t dt, (4 (l V j,β (x if, and only if, nf(c n x + d n U i,α (x and n( F(a n x + b n V j,β (x, respectively at all continuity points of U i,α (x and V j,β (x, here denotes the ea convergence, as n, means the limit as n, U,α (x = ( x α, x < 0, α > 0,, x 0, x U 2,α (x = α, x 0, α > 0, 0, x < 0,

3 Elsaah, Vishaarma and Tan / ProbStat Forum, Volume, July 208, Pages x V,β (x = β, x > 0, β > 0, ( x V, x 0, 2,β (x = β, x 0, β > 0, 0, x > 0 and U 3,0 (x = V 3,0 ( x = e x, x. In this case, e say that the distribution function F belongs to the domain of attraction of each of the limits H l(0, i,α and H u(0, j,β under linear normalization, ritten F LDA(H l(0, and F LDA(Hu(0,, respectively. i,α j,β The folloing theorem, due to Nasri-Roudsari (996 (cf. also Nasri-Roudsari and Cramer, 999, extends the above results to the symmetric generalized order statistics. Theorem.. For any symmetric generalized order statistics, let m >,, c n, a n > 0, d n, b n R, A n(m+ (x = c n(m+ x+d n(m+, B n /(m+(x = a n /(m+x+b n /(m+ and l be a fixed integer l n. Then, e have and ( An(m+ (x n n H l(m, i,α (x = t l e t dt (l U i,α(x H u(m, j,β (x = ( m+ + l if, and only if, F LDA(H l(0, and F LDA(Hu(0,, respectively. i,α j,β t m+ +l 2 e t dt, V m+ j,β (x Pantcheva (985 (cf. also Mohan and Ravi, 992 introduced a non-linear normalization for the maximal ordinary order statistics B n (x = α n x βn S(x non as the poer normalization, here S(x =, 0,, if x < 0, x = 0, x > 0, respectively. The limit distributions obtained ith non-linear normalization attract more distributions than the linear normalization. The folloing slight generalization of the results in Pantcheva (985 for the lth upper extreme ordinary order statistics under poer normalization is given by Baraat and Nigm (2002. For some suitable normalizing constants α n > 0 and β n > 0, e have Ξ (0, n l+:n (α n x βn S(x = I F(Bn(x(n l +, l Q (0, j,δ = (l ϕ j,δ (x t l e t dt, if, and only if, n ( F ( α n x βn S(x ϕ j,δ (x at all continuity points of ϕ j,δ (x, here, x 0,, x, ϕ,δ (x = (log x δ ϕ, x >, δ > 0, 2,δ (x = ( log x δ, 0 < x, δ > 0, 0, x >,, x, (log( x ϕ 3,δ (x = ( log( x δ, < x 0, δ > 0, ϕ 4,δ (x = δ, x, δ > 0, 0, x >, 0, x > 0, ϕ 5,δ (x = ϕ 5 (x = V, (x and ϕ 6,δ (x = ϕ 6 (x = V 2, (x. In this case, e say that the distribution function F belongs to the domain of attraction of Q (0, j,δ under poer normalization, ritten F PDA(Q (0, j,δ. The folloing theorem, due to Nasri-Roudsari (999, extends the above results to the symmetric generalized order statistics. Theorem.2. For any symmetric generalized order statistics, let m >,, α n, β n > 0 and l be a fixed integer l n. Then, e have ( n l+:n αn /(m+ x β n /(m+ S(x n Q (m, j,δ = ( m+ + l if, and only if, the distribution function F satisfies F PDA(Q (0, j,δ. ϕ m+ j,δ (x t m+ +l 2 e t dt,

4 Elsaah, Vishaarma and Tan / ProbStat Forum, Volume, July 208, Pages Finite mixture distribution The classical extreme value theory assumes that all the observations of a sample come from the same distribution (i.e., homogeneous population. Hoever, in real-life projects the observations come from more than one distribution (i.e., heterogeneous population ith no information to identify hich observation goes ith hich distribution. In this mixture, the ith component (distribution function of the ith subpopulation is F i (x and the mixing proportions P i > 0, i ζ are such that ζ P i =. In such cases, the distribution function is called a finite mixture distribution and is given by F(x = ζ P if i (x. The finite mixture distribution provides a natural representation of heterogeneity in many real-life data from a mixture of distributions. The finite mixture distribution has received increasing attention in recent years and has proven to be a useful approach in modeling heterogeneous data. The reader can refer to Everitt and Hand (98, Titterington et al. (985, Maclachlan and Basford (988, Lindsay (995 and Elsaah (204 for some properties and applications of finite mixture distribution. In the last fe years, much attention has been paid in exploring the potential application of the extreme value theory to the finite mixture distribution. AL-Hussaini and El-Adll (2004 studied the asymptotic distribution of the maximum ordinary order statistics under finite mixture distribution. Subsequently, Sreehari and Ravi (200 gave a closer loo at the asymptotic distribution of the maximum ordinary order statistics under finite mixture distribution ith some generalizations. Ga et al. (206 studied the asymptotic distribution of the maximum ordinary order statistics under finite mixture distribution and a statistical method to control the possible bias. Finally, Alaady et al. (206 investigated the asymptotic behavior of the appropriately linear normalized coordinateise maximum and minimum under multivariate finite mixture distribution from independent, but not obligatory identically distributed random vectors. In this paper, e extend the classical extreme value theory to the case of finite mixture distribution based on symmetric generalized order statistics under continuous strictly monotone normalization. This paper provides the possible limit distributions of upper and loer extremes symmetric generalized order statistics based on a random sample dran from a finite of mixture distributions normalized by the same continuous strictly monotonic sequence or different continuous strictly monotonic sequences. For illustration of the usage of our theoretical results, the asymptotic behaviors of mixture symmetric generalized order statistics under linear and poer normalization are studied ith some examples. 2. Mixture extreme value theory The possible non-degenerate limit distribution functions of the lth upper extreme and the lth loer extreme symmetric generalized order statistics based on a random sample dran from a finite of mixture distributions normalized by the same continuous strictly monotonic sequence, as ell as sufficient conditions for the existence of these limit distribution functions are given in Theorems 2.6 and 2.7, respectively. Theorems 2.3 and 2.4 are intended to extend the results of Theorems 2.6 and 2.7, by considering a finite of mixture distributions here the mixing distributions for the purpose of limiting distributions could be normalized by different continuous strictly monotonic sequences. The proof of the results for the lth upper extreme depends on the folloing Lemmas 2. and 2.2, hich are simple applications of Theorem.5. in Leadbetter et al. (983 (cf. also Lemma 3. and Corollary 3.2 in Nasri-Roudsari, 996 and Theorem 3 in Smirnov (952 (cf. also Lemma 3. in Baraat, 997, respectively. The proof of the results for the lth loer extreme depends on the folloing Lemmas 2.3 and 2.4, hich are simple extensions of Lemmas 2. and 2.2 respectively. Lemma 2.. For any distribution function F and non-degenerate distribution function G, let a n > 0, b n R, n N, a(t = a t, b(t = b t and t be the integral part of t. Then, the folloing statements are equivalent: F n (a n x + b n G(x for all continuity points of G.

5 Elsaah, Vishaarma and Tan / ProbStat Forum, Volume, July 208, Pages n[ F(a n x + b n ] log(g(x for all continuity points of G. F t (a(tx + b(t t G(x for all continuity points of G. t[ F(a(tx + b(t] t log(g(x for all continuity points of G. Lemma 2.2. For any distribution function F, let R be a fixed integer R N, N ( F N λ <, (x, y = y tx e t dt and σ N and ρ N N 0. Then, for large values of N e have (R (R, N ( F(x σ N I F(x (N R +, R (R (R, N ( F(x + ρ N. Lemma 2.3. For any distribution function F and non-degenerate distribution function J, let c n > 0, d n R, n N, c(t = c t, d(t = d t and t is the integral part of t. Then, the folloing statements are equivalent: [ F(c n x + d n ] n n J (x for all continuity points of J. nf(c n x + d n log(j (x for all continuity points of J. [ F(c(tx + d(t] t t J (x for all continuity points of J. tf(c(tx + d(t t log(j (x for all continuity points of J. Lemma 2.4. For any distribution function F, let N F N γ < and Θ N and Ω N N 0. Then, for large values of N e have (R (R, N F(x Θ N I F(x (R, N R + (R (R, N F(x + Ω N. Remar 2.5. The same results in Lemmas 2. and 2.3 hold ith a n x + b n and c n x + d n replaced by any continuous strictly monotonic sequence B n (x. 2.. Mixture extreme value theory via the same normalization Theorem 2.6. For any symmetric generalized order statistics from a mixture of ζ distributions F i (x ith mixture distribution function F(x = ζ P if i (x, 0 < P i < and ζ P i =, let m >, > 0, l be a fixed integer l n and L(K i (x and R(K i (x be the left and the right end points of a non-degenerate distribution function K i (x, respectively. If F i (x are normalized by the same strongly monotone continuous sequence B n (x} n= and F n i (B n (x K i (x, (5 then hen max i ζ L(K i (x < x < max i ζ R(K i (x e get n ( m+ + l Proof. Since (5 holds, from Lemma 2. and Remar 2.5 for i ζ e have ( m+ ( t ζ log(ki(xp m+ i n /(m+ ( F i (B n /(m+(x log K i (x. (6 It then follos, since ζ P i = and F(x = ζ P if i (x that F(B n /(m+(x = P i ( F i (B n /(m+(x. (7

6 Elsaah, Vishaarma and Tan / ProbStat Forum, Volume, July 208, Pages Therefore, from (6 and (7 e get ( m + + n ( F(B n /(m+(x m+ n( F(B n /(m+(x m+ [ ] m+ n /(m+ ( F(B n /(m+(x [ P i n /(m+ ( F i (B n /(m+(x ( P i log K i (x m+ ]m+. (8 From (2 and (8 and by taing F = ( F(B n /(m+(x m+, N = m+ + n and R = m+ + l in Lemma 2.2, the proof can be completed. Theorem 2.7. For any symmetric generalized order statistics from a mixture of ζ distributions F i (x ith mixture distribution function F(x = ζ P if i (x, 0 < P i < and ζ P i =, let m >, > 0, l be a fixed integer l n and L(E i (x and R(E i (x be the left and the right end points of a non-degenerate distribution function E i (x, respectively. If F i (x are normalized by the same strongly monotone continuous sequence A n (x} n= and ( F i (A n (x n n E i (x, (9 then hen max i ζ L(E i (x < x < max i ζ R(E i (x e have (A n(m+ (x (l ζ log(ei(x P i t l e t dt. Proof. Since (9 holds, from Lemma 2.3 and Remar 2.5 e get n(m + F i (A n(m+ (x log E i (x. (0 It then follos, since ζ P i = and F(A n(m+ (x = ζ P if i (A n(m+ (x that F(A n(m+ (x = ( P i Fi (A n(m+ (x. ( From Maclaurin series, e have ( F(A n(m+ (x m+ = (m + F(A n(m+ (x( + ( (m + F(A n(m+ (x, here ( 0. Therefore, from (0, ( and (2 e get ( ( m + + n ( F(An(m+(x m+ n ( ( F(A n(m+(x m+ n(m + F(A n(m+ (x P i n(m + F i (A n(m+ (x (2 P i log E i (x. (3 From ( and (3 and by taing F = ( F(A n(m+ (x m+, N = m+ + n and R = l in Lemma 2.4, the proof can be completed.

7 Elsaah, Vishaarma and Tan / ProbStat Forum, Volume, July 208, Pages Corollary 2.8. From Theorems 2.6 and 2.7, e have Ξ (0, n l+:n (B n(x and 0, if x max i ζ L(K i (x, n, if x max i ζ R(K i (x and Ξ (0, (A n(x and (A n(m+ (x 0, if x max i ζ L(E i (x,, if x max i ζ R(E i (x. Corollary 2.9. From Theorems 2.6 and 2.7, e have the folloing special cases. It is orth noting that, the special cases (iii in (I and (II can be found in AL-Hussaini and El-Adll (2004 and Sreehari and Ravi (200. (I For ordinary order statistics, i.e., m = 0 and =, e get (i Ξ (0, n l+:n (B n(x (ii Ξ (0, (A n(x (iii Ξ (0, n:n (B n (x (iv Ξ (0, :n (A n(x ζ (l (l log(ki(x P i tl e t dt. ζ ζ (K i(x Pi. ζ (E i(x Pi. log(ei(x P i tl e t dt. (II If the observations come from the same distribution, i.e., K i (x = K(x i and E i (x = E(x i, then (i (ii Ξ (0, (iii Ξ (0, n:n (B n (x ( (A n(x and (A n(m+ (x K(x. (iv Ξ (0, :n (A n(x and :n (A n(m+(x +l ( m+ (log K(x t m+ m+ (l E(x. log E(x tl e t dt. Corollary 2.0. For any symmetric generalized order statistics from a mixture of ζ distributions normalized by the same linear sequence B n (x = a n x + b n, let m >,, a n > 0, b n R, l be a fixed integer l n and L(K i (x and R(K i (x be the left and the right end points of a non-degenerate distribution function K i (x, respectively. Then, hen max i ζ L(K i (x < x < max i ζ R(K i (x, e get (I Let F(x = PF (x+( PF 2 (x, here 0 < P <, F LDA(H u(0,,β (x and F 2 LDA(H u(0, 2,β (x. Then, e get F(x LDA(H u(0,,β (x. That is, for x > 0 e get ( +l (Px β t m+ m+ (II Let F(x = PF (x+( PF 2 (x, here 0 < P <, F LDA(H u(0,,β (x and F 2 LDA(H u(0, 3,β (x. Then for x > 0, e get the folloing non-degenerate distribution ( +l (P(x β e x +e x t m+ m+ (III Let F(x = PF (x+( PF 2 (x, here 0 < P <, F LDA(H u(0, 2,β (x and F 2 LDA(H u(0, 3,β (x. Then, e have n l+:n (B n /(m+(x +l (e x +P(( x β e x t m+ m+ +l (( Pe x t m+ m+ ( ( m+ +l 2 e t dt, x 0, m+ +l 2 e t dt, x > 0.

8 Elsaah, Vishaarma and Tan / ProbStat Forum, Volume, July 208, Pages 04 6 (IV Let F(x = P F (x + P 2 F 2 (x + P 3 F 3 (x, here P + P 2 + P 3 =, 0 < P, P 2, P 3 <, F (m, LDA( H,β (x, F 2 LDA(H u(0, 2,β (x and F 3 LDA(H u(0, 3,β (x. Then for x > 0, ( +l (P m+ x β +P 3e x t m+ Proof. The proof is obvious from Theorems. and 2.6 ith simple algebra. We ill provide the proof for the first case (I and the proof for the other cases ill be by the same technique. Let F(x = PF (x + ( PF 2 (x, here 0 < P <, F LDA(H u(0,,β (x and F 2 LDA(H u(0, 2,β (x. Then, from Theorems. and 2.6 e get n ( m+ + l ( m+ ( PV,β (x ( PV 2,β (x m+ t From (4, e get Thus, e get PV,β (x + ( PV 2,β (x = Px β, x > 0, β > 0,, x 0, PV,β (x + ( PV 2,β (x = ( P( x + β, x 0, β > 0, 0, x > 0 Px β, x > 0, β > 0,, x 0. Corollary 2.. For any symmetric generalized order statistics from a mixture of ζ distributions normalized by the same strongly monotone continuous poer sequence B n (x = α n x βn S(x, let m >,, α n, β n > 0, l be a fixed integer l n and L(K i (x and R(K i (x be the left and the right end points of a nondegenerate distribution function K i (x, respectively. Then, hen max i ζ L(K i (x < x < max i ζ R(K i (x, e get (I Let F(x = 5 P if i (x, here 5 P i =, 0 < P i <, F PDA(Q (0, PDA(Q (0, t,δ (x, t 2, 3, 4, 6}, 2 j 5. Then for x >, e get ( +l (P m+ (log x δ t m+,δ (x and F j (II Let F(x = 6 P if i (x, here 6 P i =, 0 < P i <, F PDA(Q (0,,δ (x, F 5 PDA(Q (0, 5,δ (x and F j PDA(Q (0, t,δ (x, t 2, 3, 4, 6}, j = 2, 3, 4, 6. Then for x >, e get ( (P(log +l x δ +P 5 m+ x m+ t (III Let F(x = 4 P if i (x, here 4 P i =, 0 < P i <, F 2 PDA(Q (0, PDA(Q (0, t,δ (x, t 3, 4, 6}, j =, 3, 4. Then for 0 < x, e get ( +l (( δ P m+ 2(log x δ t m+ 2,δ (x and F j (IV Let F(x = 5 P if i (x, here 5 P i =, 0 < P i <, F 2 PDA(Q (0, 2,δ (x, F 5 PDA(Q (0, 5,δ (x and F j PDA(Q (0, t,δ (x, t 3, 4, 6}, j =, 3, 4. Then, n ( +l m+ (P 2( log x δ + P 5 x m+ t m+ +l 2 e t dt, 0 < x <, ( +l m+ ( P 5 x m+ t m+ +l 2 e t dt, x >.

9 Elsaah, Vishaarma and Tan / ProbStat Forum, Volume, July 208, Pages (V Let F(x = 4 P if i (x, here 4 P i =, 0 < P i <, F PDA(Q (0, PDA(Q (0, t,δ (x, t 3, 4, 6}, j = 2, 3, 4. Then for x > 0, e get n ( +l m+ ( P x m+ t 5,δ (x and F j Proof. The proof is obvious from Theorems.2 and 2.6 by the same technique as that in the proof of Corollary 2.0. Remar 2.2. If Ξ (0, ( +l z m+ t m+ +l 2 e t dt, then n l+:n (B n(x e z. Therefore, the discussion in Corollaries 2.0 and 2. can be immediately given for the ordinary order statistics Mixture extreme value theory via different normalizations Theorem 2.3. For any symmetric generalized order statistics from a mixture of ζ distributions F i (x ith mixture distribution function F(x = ζ P ζ if i (x, 0 < P i <, P i =, let m >, > 0, l be a fixed integer l n and L(K i (x and R(K i (x be the left and the right end points of a nondegenerate distribution function K i (x, respectively. Suppose F i (x are normalized by different strongly monotone continuous sequences B i,n (x} n= such that, for i ζ, the folloing conditions hold (C u F n i (B i,n(x K i (x, 0 < K i (x <, (C u 2 n[f i (B n (x F i (B i,n (x] η i (x, 0 < η i (x <, for some strongly monotone continuous sequence B n (x} n=. Then, hen max i ζ L(K i (x < x < max i ζ R(K i (x, e have n ( m+ + l ( t ζ [log(k i(x P i +P iη i(x] m+ Proof. Since F(x = ζ P if i (x, e get F(B n /(m+(x = P i F i (B i,n /(m+(x + [ P i Fi (B n /(m+(x F i (B i,n /(m+(x ] ζ = P ζ F ζ (B ζ,n /(m+(x + P i F i (B i,n /(m+(x + [ P i Fi (B n /(m+(x F i (B i,n /(m+(x ]. (4 Since P ζ = ζ P i, from (4 e have ζ [ F(B n /(m+(x = F ζ (B ζ,n /(m+(x P i Fζ (B ζ,n /(m+(x F i (B i,n /(m+(x ] + [ P i Fi (B n /(m+(x F i (B i,n /(m+(x ]. (5

10 Elsaah, Vishaarma and Tan / ProbStat Forum, Volume, July 208, Pages Thus, from (5 e get n /(m+ ( F(B n /(m+(x = n /(m+ ( F ζ (B ζ,n /(m+(x P i n /(m+ [ F i (B n /(m+(x F i (B i,n /(m+(x ] P i [n /(m+ ( F i (B i,n /(m+(x ζ + ] n /(m+ ( F ζ (B ζ,n /(m+(x. (6 Suppose that (C u is satisfied. Then, from Lemma 2. and Remar 2.5 for i =, 2,..., ζ e have n( F i (B n (x log K i (x. Therefore, (C u and (C2 u become and n /(m+ ( F i (B n /(m+(x log K i (x (7 n /(m+ [F i (B n /(m+(x F i (B i,n /(m+(x] η i (x, (8 respectively. Then, from (6, (7 and (8 e get n /(m+ ( F(B n /(m+(x log K ζ (x It then follos, since ζ P i = P ζ, that n /(m+ ( F(B n /(m+(x ζ + P i η i (x P i ( log K i (x + log K ζ (x. (9 P i (log K i (x + η i (x. (20 Therefore, from (20 e get ( m + + n ( F(B n /(m+(x m+ n [( F(B n /(m+(x] m+ [ ] m+ n /(m+ ( F(B n /(m+(x [ P i (log K i (x + η i (x From (2 and (2 and by taing F = ( F(B n /(m+(x m+, N = m+ + n and R = m+ + l in Lemma 2.2, the proof can be completed. Theorem 2.4. For any symmetric generalized order statistics from a mixture of ζ distributions F i (x ith mixture distribution function F(x = ζ P ζ if i (x, 0 < P i <, P i =, let m >, > 0, l be a fixed integer l n and L(E i (x and R(E i (x be the left and the right end points of a non-degenerate distribution function E i (x, respectively. Suppose F i (x are normalized by different strongly monotone continuous sequences A i,n (x} n= such that, for i ζ, the folloing conditions hold (C l ( F i (A i,n (x n n E i (x, 0 < E i (x <, ]m+. (2

11 Elsaah, Vishaarma and Tan / ProbStat Forum, Volume, July 208, Pages (C l 2 n[f i (A n (x F i (A i,n (x] µ i (x, 0 < µ i (x <, for some strongly monotone continuous sequence A n (x} n=. Then, hen max i ζ L(E i (x < x < max i ζ R(E i (x, e have (A n(m+ (x (l ζ [log(e i(x P i P iµ i(x] tl e t dt. Proof. The proof is similar to the proof of Theorems 2.7 and 2.3, except the obvious changes. Corollary 2.5. From Theorems 2.3 and 2.4, e have the folloing special cases (I For ordinary order statistics, i.e., m = 0 and =, e get (i Ξ (0, n l+:n (B n(x (ii Ξ (0, (A n(x ζ (l (l (iii Ξ (0, n:n (B n (x = F n (B n (x (iv Ξ (0, :n (A n(x [log(k i(x P i P iη i(x] tl e t dt. [log(ei(x P i P iµ i(x] tl e t dt. ζ [K i(x exp(η i (x] Pi. ζ [E i(x exp(µ i (x] Pi. (II If the observations come from the same distribution, i.e., K i (x = K(x, η i (x = η(x, E i (x = E(x and µ i (x = µ(x for i ζ, then e have (i ( (ii (A n(m+ (x and Ξ (0, (A n(x (iii Ξ (0, n:n (B n (x K(x exp(η(x. (iv :n (A n(m+(x and Ξ (0, :n (A n(x +l ( m+ (log K(x+η(x t m+ m+ (l log E(x µ(x tl e t dt. E(x exp(µ(x. 3. Applications In this section some illustrative examples of the most practically important distributions are obtained, hich lend further support to our theoretical results. Example 3.. (logistic, Exponential( and Gumbel distributions. Let X, X 2,..., X n be a random sample dran from a mixed population hose distribution function F(x = P F (x + P 2 F 2 (x + P 3 F 3 (x, here P + P 2 + P 3 =, 0 < P i <, i =, 2, 3, F (x = +e, < x <, F x 2 (x = e x, x > 0 and F 3 (x = exp( e x, < x <. One can choose the common linear normalization, B n (x = x+log n. Then B n /(m+ = x+ m+ log n. It is not difficult to sho that n( F i(x+log n e x, or, equivalently, Fi n (x + log n n exp( e x, i =, 2, 3. Then, e have n l+:n (x + log n m + n ( m+ + l and Ξ (0, n:n (x + log n exp( e x. e x(m+ t m+ +l 2 e t dt Example 3.2. (Poer( and Exponential( distributions. Let X, X 2,..., X n be a random sample dran from a mixed population hose distribution function F(x = PF (x + ( PF 2 (x, here 0 < P <, F (x = x, 0 < x < and F 2 (x = e x, x > 0. One can choose the common linear normalization, A n (x = x n. Then, e get A x n(m+(x = n(m+. It is not difficult to sho that nf i( x n x, i =, 2.

12 Elsaah, Vishaarma and Tan / ProbStat Forum, Volume, July 208, Pages ( or, equivalently, ( F i ( x n n n e x, i =, 2. Then, e have x n(m+ and Ξ (0, ( x n (l x tl e t dt and Ξ (0, ( x :n n n e x. Example 3.3. Let X, X 2,..., X n be a random sample dran from a mixed population hose distribution function F(x = PF (x + ( PF 2 (x, here 0 < P <, F (x = exp[ ( log x α ], 0 < x 0, x 0,, x >, exp[ (log( x and F 2 (x = α ], x, One can choose the common poer normalization, B, x >. n (x = x n α S(x. Then B n /(m+(x = x n α(m+ S(x. It is not difficult to sho that ( ] n [F i x n /α S(x n F i (x, i =, 2. Then, e have n l+:n ( x n α(m+ S(x ( m+ + l ( and Ξ (0, n:n x n α S(x ( m+ (log([f (x] P [F 2(x] P m+ t m+ +l 2 e t dt [F (x] P [F 2 (x] P. Example 3.4. (Pareto and Frechet distributions. Let X, X 2,..., X n be a random sample dran from a mixed population hose distribution function F(x = PF (x + ( PF 2 (x, here 0 < P < 0, x <, exp[ x, F (x = x α and F, x, α > 0 2 (x = α ], x 0, α > 0 One can choose the 0, x < 0. common poer normalization, B n (x = (nx /α. Then, e get B n /(m+(x = (n m+ x /α. It is not difficult to sho that n( F i ((nx /α n x, i =, 2. Then, e get ( n l+:n (n m+ x /α n ( m+ + l ( and Ξ (0, n:n (nx /α = n e /x. x (m+ t m+ +l 2 e t dt Similarly, one can choose the common linear normalization, B n (x = n /α x. Then, B n /(m+(x = n /α(m+ x. It is not difficult to sho that n( F i (n /α x x α, i =, 2. Then, ( n l+:n n /α(m+ x n ( m+ + l ( and Ξ n:n (0, n /α x = n e x α. x α(m+ t m+ +l 2 e t dt Remar 3.. It is orth mentioning that at this stage e are unable to find a suitable example here the normalizing constants are different to support the assumptions (C u 2 and (C l 2 of Theorems 2.3 and 2.4, respectively. This ill be investigated in our future or. Acnoledgments The authors greatly appreciate the corrections and helpful suggestions by the referee that significantly improved the paper. Elsaah also than Prof. Kai-Tai Fang for his guidance and ind support during this or.

13 References Elsaah, Vishaarma and Tan / ProbStat Forum, Volume, July 208, Pages [] Alaady, M.A, Elsaah, A.M, Hu, J. and Qin, H., 206. Asymptotic behavior of non-identical multivariate mixture. ProbStat Forum, 09, July 206, [2] AL-Hussaini, E.K. and El-Adll, E.M., Asymptotic distribution of normalized maximum under finite mixture models. Statist. Probab. Lett. 70, [3] Balema, A.A. and de Haan, L., 972. On R. von Mises condition for the domain of attraction of exp( e x. Ann. Math. Stat. 43, [4] Baraat, H.M., 997. Asymptotic properties of bivariate random extremes. J. Statist. Plann. Inference 6, [5] Baraat, H.M and Nigm, E.M, Extreme order statistics under poer normalization and random sample size. Kuait J. Sci. Eng. 29(, [6] Elsaah, A.M., 204. Generalized Order Statistics Under Finite Mixture Models: Asymptotic Theory and Applications. Lap Lambert Academic Publishing, Riga. [7] Everitt, B.S. and Hand, D.J., 98. Finite Mixture Distribution. Chapman and Hall, London. [8] Gnedeno, B.V., 943. Sur la distribution limite du terme maximum dune serie al eatoire. Ann. Math. 44, [9] Ga, W., Goo, H., Choi, Y.H. and Ahn, J.Y., 206. Extreme value theory in mixture distributions and a statistical method to control the possible bias. J. Korean Statist. Soc. 45, [0] Kamps, U., 995. A Concept of Generalized Order Statistics. Teubner, Stuttgart. Update. 3, Wiley, Ne Yor. [] Leadbetter, M.R., Lindgren, G. and Rootzen, H., 983, Extremes and related properties of random sequences and processes. Springer, Ne Yor. [2] Lindsay, B.G., 995. Mixture Models: Theory, Geometry and Applications. The Institute of Mathematical Statistics, Hayard, CA. [3] Maclachlan, G.J. and Basford, K.E., 988. Mixture Models: Applications to Clustering. Marcel Deer, Ne Yor. [4] Mohan, N.R. and Ravi, S., 992. Max domains of attraction of univariate and multivariate p-max stable las. Theory Probab. Appl. 37, [5] Nasri-Roudsari, D., 996. Extreme value theory of generalized order statistics. J. Statist. Plann. Inference 55, [6] Nasri-Roudsari, D., 999. Limit distributions of generalized order statistics under poer normalization. Commun. Statist.- Theory Meth. 28(6, [7] Nasri-Roudsari, D. and Cramer, E., 999. On the convergence rates of extreme generalized order statistics. Extremes 2, [8] Pantcheva, E., 985. Limit theorems for extreme order statistics under nonlinear normalization. Lecture Notes in Mathematics, vol. 55, pp [9] Smirnov, N. V. 949, 952. Limit distributions for the terms of a variational series. Original Russian in Trudy Math. Inst. Stelov., 25 (949, -6. [20] Sreehari, M. and Ravi, S., 200. On extremes of mixture of distributions. Metria 7, [2] Titterington, D.M., Smith, A.F.M. and Maov, U.E., 985. Statistical Analysis of Finite Mixture Distributions, Wiley. Ne Yor.

Limit Distributions of Extreme Order Statistics under Power Normalization and Random Index

Limit Distributions of Extreme Order Statistics under Power Normalization and Random Index Limit Distributions of Extreme Order tatistics under Power Normalization and Random Index Zuoxiang Peng, Qin Jiang & aralees Nadarajah First version: 3 December 2 Research Report No. 2, 2, Probability

More information

A Note on Tail Behaviour of Distributions. the max domain of attraction of the Frechét / Weibull law under power normalization

A Note on Tail Behaviour of Distributions. the max domain of attraction of the Frechét / Weibull law under power normalization ProbStat Forum, Volume 03, January 2010, Pages 01-10 ISSN 0974-3235 A Note on Tail Behaviour of Distributions in the Max Domain of Attraction of the Frechét/ Weibull Law under Power Normalization S.Ravi

More information

Limit Laws for Maxima of Functions of Independent Non-identically Distributed Random Variables

Limit Laws for Maxima of Functions of Independent Non-identically Distributed Random Variables ProbStat Forum, Volume 07, April 2014, Pages 26 38 ISSN 0974-3235 ProbStat Forum is an e-journal. For details please visit www.probstat.org.in Limit Laws for Maxima of Functions of Independent Non-identically

More information

Stochastic Comparisons of Order Statistics from Generalized Normal Distributions

Stochastic Comparisons of Order Statistics from Generalized Normal Distributions A^VÇÚO 1 33 ò 1 6 Ï 2017 c 12 Chinese Journal of Applied Probability and Statistics Dec. 2017 Vol. 33 No. 6 pp. 591-607 doi: 10.3969/j.issn.1001-4268.2017.06.004 Stochastic Comparisons of Order Statistics

More information

Chapter 2 Asymptotics

Chapter 2 Asymptotics Chapter Asymptotics.1 Asymptotic Behavior of Student s Pdf Proposition.1 For each x R d,asν, f ν,,a x g a, x..1 Proof Let a = 0. Using the well-known formula that Ɣz = π z e z z z 1 + O 1, as z,. z we

More information

Enhancing Generalization Capability of SVM Classifiers with Feature Weight Adjustment

Enhancing Generalization Capability of SVM Classifiers with Feature Weight Adjustment Enhancing Generalization Capability of SVM Classifiers ith Feature Weight Adjustment Xizhao Wang and Qiang He College of Mathematics and Computer Science, Hebei University, Baoding 07002, Hebei, China

More information

On discrete distributions with gaps having ALM property

On discrete distributions with gaps having ALM property ProbStat Forum, Volume 05, April 202, Pages 32 37 ISSN 0974-3235 ProbStat Forum is an e-journal. For details please visit www.probstat.org.in On discrete distributions with gaps having ALM property E.

More information

Bivariate Uniqueness in the Logistic Recursive Distributional Equation

Bivariate Uniqueness in the Logistic Recursive Distributional Equation Bivariate Uniqueness in the Logistic Recursive Distributional Equation Antar Bandyopadhyay Technical Report # 629 University of California Department of Statistics 367 Evans Hall # 3860 Berkeley CA 94720-3860

More information

Precise Asymptotics of Generalized Stochastic Order Statistics for Extreme Value Distributions

Precise Asymptotics of Generalized Stochastic Order Statistics for Extreme Value Distributions Applied Mathematical Sciences, Vol. 5, 2011, no. 22, 1089-1102 Precise Asymptotics of Generalied Stochastic Order Statistics for Extreme Value Distributions Rea Hashemi and Molood Abdollahi Department

More information

Extreme Value Analysis and Spatial Extremes

Extreme Value Analysis and Spatial Extremes Extreme Value Analysis and Department of Statistics Purdue University 11/07/2013 Outline Motivation 1 Motivation 2 Extreme Value Theorem and 3 Bayesian Hierarchical Models Copula Models Max-stable Models

More information

Reliability of Coherent Systems with Dependent Component Lifetimes

Reliability of Coherent Systems with Dependent Component Lifetimes Reliability of Coherent Systems with Dependent Component Lifetimes M. Burkschat Abstract In reliability theory, coherent systems represent a classical framework for describing the structure of technical

More information

Mi-Hwa Ko. t=1 Z t is true. j=0

Mi-Hwa Ko. t=1 Z t is true. j=0 Commun. Korean Math. Soc. 21 (2006), No. 4, pp. 779 786 FUNCTIONAL CENTRAL LIMIT THEOREMS FOR MULTIVARIATE LINEAR PROCESSES GENERATED BY DEPENDENT RANDOM VECTORS Mi-Hwa Ko Abstract. Let X t be an m-dimensional

More information

Maximum Likelihood and Bayes Estimations under Generalized Order Statistics from Generalized Exponential Distribution

Maximum Likelihood and Bayes Estimations under Generalized Order Statistics from Generalized Exponential Distribution Applied Mathematical Sciences, Vol. 6, 2012, no. 49, 2431-2444 Maximum Likelihood and Bayes Estimations under Generalized Order Statistics from Generalized Exponential Distribution Saieed F. Ateya Mathematics

More information

Endogeny for the Logistic Recursive Distributional Equation

Endogeny for the Logistic Recursive Distributional Equation Zeitschrift für Analysis und ihre Anendungen Journal for Analysis and its Applications Volume 30 20, 237 25 DOI: 0.47/ZAA/433 c European Mathematical Society Endogeny for the Logistic Recursive Distributional

More information

Introduction A basic result from classical univariate extreme value theory is expressed by the Fisher-Tippett theorem. It states that the limit distri

Introduction A basic result from classical univariate extreme value theory is expressed by the Fisher-Tippett theorem. It states that the limit distri The dependence function for bivariate extreme value distributions { a systematic approach Claudia Kluppelberg Angelika May October 2, 998 Abstract In this paper e classify the existing bivariate models

More information

CHARACTERIZATIONS OF POWER DISTRIBUTIONS VIA MOMENTS OF ORDER STATISTICS AND RECORD VALUES

CHARACTERIZATIONS OF POWER DISTRIBUTIONS VIA MOMENTS OF ORDER STATISTICS AND RECORD VALUES APPLICATIOES MATHEMATICAE 26,41999, pp. 467 475 Z. GRUDZIEŃ and D. SZYALLublin CHARACTERIZATIOS OF POWER DISTRIBUTIOS VIA MOMETS OF ORDER STATISTICS AD RECORD VALUES Abstract. Power distributions can be

More information

On the chromatic Ext 0 (Mn 1 1 ) on Γ(m + 1) for an odd prime

On the chromatic Ext 0 (Mn 1 1 ) on Γ(m + 1) for an odd prime On the chromatic Ext 0 Mn 1 1 on Γm + 1 for an odd prime Rié Kitahama and Katsumi Shimomura Received Xxx 00, 0000 Abstract. Let Mn 1 1 denote the coernel of the localization map BP /I v 1 n 1 BP /I, here

More information

Estimation Under Multivariate Inverse Weibull Distribution

Estimation Under Multivariate Inverse Weibull Distribution Global Journal of Pure and Applied Mathematics. ISSN 097-768 Volume, Number 8 (07), pp. 4-4 Research India Publications http://www.ripublication.com Estimation Under Multivariate Inverse Weibull Distribution

More information

MOMENT CONVERGENCE RATES OF LIL FOR NEGATIVELY ASSOCIATED SEQUENCES

MOMENT CONVERGENCE RATES OF LIL FOR NEGATIVELY ASSOCIATED SEQUENCES J. Korean Math. Soc. 47 1, No., pp. 63 75 DOI 1.4134/JKMS.1.47..63 MOMENT CONVERGENCE RATES OF LIL FOR NEGATIVELY ASSOCIATED SEQUENCES Ke-Ang Fu Li-Hua Hu Abstract. Let X n ; n 1 be a strictly stationary

More information

Homework 6 Solutions

Homework 6 Solutions Homeork 6 Solutions Igor Yanovsky (Math 151B TA) Section 114, Problem 1: For the boundary-value problem y (y ) y + log x, 1 x, y(1) 0, y() log, (1) rite the nonlinear system and formulas for Neton s method

More information

AN ASYMPTOTICALLY UNBIASED MOMENT ESTIMATOR OF A NEGATIVE EXTREME VALUE INDEX. Departamento de Matemática. Abstract

AN ASYMPTOTICALLY UNBIASED MOMENT ESTIMATOR OF A NEGATIVE EXTREME VALUE INDEX. Departamento de Matemática. Abstract AN ASYMPTOTICALLY UNBIASED ENT ESTIMATOR OF A NEGATIVE EXTREME VALUE INDEX Frederico Caeiro Departamento de Matemática Faculdade de Ciências e Tecnologia Universidade Nova de Lisboa 2829 516 Caparica,

More information

The Complementary Exponential-Geometric Distribution Based On Generalized Order Statistics

The Complementary Exponential-Geometric Distribution Based On Generalized Order Statistics Applied Mathematics E-Notes, 152015, 287-303 c ISSN 1607-2510 Available free at mirror sites of http://www.math.nthu.edu.tw/ amen/ The Complementary Exponential-Geometric Distribution Based On Generalized

More information

Non linear functionals preserving normal distribution and their asymptotic normality

Non linear functionals preserving normal distribution and their asymptotic normality Armenian Journal of Mathematics Volume 10, Number 5, 018, 1 0 Non linear functionals preserving normal distribution and their asymptotic normality L A Khachatryan and B S Nahapetian Abstract We introduce

More information

Initial Coefficient Bounds for a General Class of Bi-Univalent Functions

Initial Coefficient Bounds for a General Class of Bi-Univalent Functions Filomat 29:6 (2015), 1259 1267 DOI 10.2298/FIL1506259O Published by Faculty of Sciences Mathematics, University of Niš, Serbia Available at: http://.pmf.ni.ac.rs/filomat Initial Coefficient Bounds for

More information

Journal of Inequalities in Pure and Applied Mathematics

Journal of Inequalities in Pure and Applied Mathematics Journal of Inequalities in Pure and Applied Mathematics ON A HYBRID FAMILY OF SUMMATION INTEGRAL TYPE OPERATORS VIJAY GUPTA AND ESRA ERKUŞ School of Applied Sciences Netaji Subhas Institute of Technology

More information

arxiv: v1 [cs.it] 12 Jun 2016

arxiv: v1 [cs.it] 12 Jun 2016 New Permutation Trinomials From Niho Exponents over Finite Fields with Even Characteristic arxiv:606.03768v [cs.it] 2 Jun 206 Nian Li and Tor Helleseth Abstract In this paper, a class of permutation trinomials

More information

Asymptotic behavior for sums of non-identically distributed random variables

Asymptotic behavior for sums of non-identically distributed random variables Appl. Math. J. Chinese Univ. 2019, 34(1: 45-54 Asymptotic behavior for sums of non-identically distributed random variables YU Chang-jun 1 CHENG Dong-ya 2,3 Abstract. For any given positive integer m,

More information

Tamsui Oxford Journal of Information and Mathematical Sciences 29(2) (2013) Aletheia University

Tamsui Oxford Journal of Information and Mathematical Sciences 29(2) (2013) Aletheia University Tamsui Oxford Journal of Information and Mathematical Sciences 292 2013 219-238 Aletheia University Relations for Marginal and Joint Moment Generating Functions of Extended Type I Generalized Logistic

More information

Limits at Infinity. Horizontal Asymptotes. Definition (Limits at Infinity) Horizontal Asymptotes

Limits at Infinity. Horizontal Asymptotes. Definition (Limits at Infinity) Horizontal Asymptotes Limits at Infinity If a function f has a domain that is unbounded, that is, one of the endpoints of its domain is ±, we can determine the long term behavior of the function using a it at infinity. Definition

More information

Z n -GRADED POLYNOMIAL IDENTITIES OF THE FULL MATRIX ALGEBRA OF ORDER n

Z n -GRADED POLYNOMIAL IDENTITIES OF THE FULL MATRIX ALGEBRA OF ORDER n PROCEEDINGS OF THE AMERICAN MATHEMATICAL SOCIETY Volume 127, Number 12, Pages 3517 3524 S 0002-9939(99)04986-2 Article electronically published on May 13, 1999 Z n -GRADED POLYNOMIAL IDENTITIES OF THE

More information

Non Uniform Bounds on Geometric Approximation Via Stein s Method and w-functions

Non Uniform Bounds on Geometric Approximation Via Stein s Method and w-functions Communications in Statistics Theory and Methods, 40: 45 58, 20 Copyright Taylor & Francis Group, LLC ISSN: 036-0926 print/532-45x online DOI: 0.080/036092090337778 Non Uniform Bounds on Geometric Approximation

More information

Asymptotic statistics using the Functional Delta Method

Asymptotic statistics using the Functional Delta Method Quantiles, Order Statistics and L-Statsitics TU Kaiserslautern 15. Februar 2015 Motivation Functional The delta method introduced in chapter 3 is an useful technique to turn the weak convergence of random

More information

Method of Conditional Moments Based on Incomplete Data

Method of Conditional Moments Based on Incomplete Data , ISSN 0974-570X (Online, ISSN 0974-5718 (Print, Vol. 20; Issue No. 3; Year 2013, Copyright 2013 by CESER Publications Method of Conditional Moments Based on Incomplete Data Yan Lu 1 and Naisheng Wang

More information

Bayesian Inference for Clustered Extremes

Bayesian Inference for Clustered Extremes Newcastle University, Newcastle-upon-Tyne, U.K. lee.fawcett@ncl.ac.uk 20th TIES Conference: Bologna, Italy, July 2009 Structure of this talk 1. Motivation and background 2. Review of existing methods Limitations/difficulties

More information

Solution of the Higher Order Cauchy Difference Equation on Free Groups

Solution of the Higher Order Cauchy Difference Equation on Free Groups Global Journal of Pure and Applied Mathematics. ISSN 0973-1768 Volume 13, Number 8 (2017, pp. 4141 4151 Research India Publications http://www.ripublication.com/gjpam.htm Solution of the Higher Order Cauchy

More information

Almost Sure Convergence of the General Jamison Weighted Sum of B-Valued Random Variables

Almost Sure Convergence of the General Jamison Weighted Sum of B-Valued Random Variables Acta Mathematica Sinica, English Series Feb., 24, Vol.2, No., pp. 8 92 Almost Sure Convergence of the General Jamison Weighted Sum of B-Valued Random Variables Chun SU Tie Jun TONG Department of Statistics

More information

Math 576: Quantitative Risk Management

Math 576: Quantitative Risk Management Math 576: Quantitative Risk Management Haijun Li lih@math.wsu.edu Department of Mathematics Washington State University Week 11 Haijun Li Math 576: Quantitative Risk Management Week 11 1 / 21 Outline 1

More information

Chapter 6. Order Statistics and Quantiles. 6.1 Extreme Order Statistics

Chapter 6. Order Statistics and Quantiles. 6.1 Extreme Order Statistics Chapter 6 Order Statistics and Quantiles 61 Extreme Order Statistics Suppose we have a finite sample X 1,, X n Conditional on this sample, we define the values X 1),, X n) to be a permutation of X 1,,

More information

High-order Newton-type iterative methods with memory for solving nonlinear equations

High-order Newton-type iterative methods with memory for solving nonlinear equations MATHEMATICAL COMMUNICATIONS 9 Math. Commun. 9(4), 9 9 High-order Newton-type iterative methods with memory for solving nonlinear equations Xiaofeng Wang, and Tie Zhang School of Mathematics and Physics,

More information

Modelling Dropouts by Conditional Distribution, a Copula-Based Approach

Modelling Dropouts by Conditional Distribution, a Copula-Based Approach The 8th Tartu Conference on MULTIVARIATE STATISTICS, The 6th Conference on MULTIVARIATE DISTRIBUTIONS with Fixed Marginals Modelling Dropouts by Conditional Distribution, a Copula-Based Approach Ene Käärik

More information

Renewal Theory and Geometric Infinite Divisibility

Renewal Theory and Geometric Infinite Divisibility ProbStat Models, 2, May-2003, Pages 1-8 Department of Statistics Valiavilakom Prajyoti Niketan College Ookode Pudukkad, Trichur 680 301 Trivandrum 695 020 India. India. esandhya@hotmail.com Received in

More information

Faber polynomial coefficient bounds for a subclass of bi-univalent functions

Faber polynomial coefficient bounds for a subclass of bi-univalent functions Stud. Univ. Babeş-Bolyai Math. 6(06), No., 37 Faber polynomial coefficient bounds for a subclass of bi-univalent functions Şahsene Altınkaya Sibel Yalçın Abstract. In this ork, considering a general subclass

More information

COMPLETE QTH MOMENT CONVERGENCE OF WEIGHTED SUMS FOR ARRAYS OF ROW-WISE EXTENDED NEGATIVELY DEPENDENT RANDOM VARIABLES

COMPLETE QTH MOMENT CONVERGENCE OF WEIGHTED SUMS FOR ARRAYS OF ROW-WISE EXTENDED NEGATIVELY DEPENDENT RANDOM VARIABLES Hacettepe Journal of Mathematics and Statistics Volume 43 2 204, 245 87 COMPLETE QTH MOMENT CONVERGENCE OF WEIGHTED SUMS FOR ARRAYS OF ROW-WISE EXTENDED NEGATIVELY DEPENDENT RANDOM VARIABLES M. L. Guo

More information

A NOTE ON SECOND ORDER CONDITIONS IN EXTREME VALUE THEORY: LINKING GENERAL AND HEAVY TAIL CONDITIONS

A NOTE ON SECOND ORDER CONDITIONS IN EXTREME VALUE THEORY: LINKING GENERAL AND HEAVY TAIL CONDITIONS REVSTAT Statistical Journal Volume 5, Number 3, November 2007, 285 304 A NOTE ON SECOND ORDER CONDITIONS IN EXTREME VALUE THEORY: LINKING GENERAL AND HEAVY TAIL CONDITIONS Authors: M. Isabel Fraga Alves

More information

The coincidence Nielsen number for maps into real projective spaces

The coincidence Nielsen number for maps into real projective spaces F U N D A M E N T A MATHEMATICAE 140 (1992) The coincidence Nielsen number for maps into real projective spaces by Jerzy J e z i e r s k i (Warszawa) Abstract. We give an algorithm to compute the coincidence

More information

Random Infinite Divisibility on Z + and Generalized INAR(1) Models

Random Infinite Divisibility on Z + and Generalized INAR(1) Models ProbStat Forum, Volume 03, July 2010, Pages 108-117 ISSN 0974-3235 Random Infinite Divisibility on Z + and Generalized INAR(1) Models Satheesh S NEELOLPALAM, S. N. Park Road Trichur - 680 004, India. ssatheesh1963@yahoo.co.in

More information

J. Cwik and J. Koronacki. Institute of Computer Science, Polish Academy of Sciences. to appear in. Computational Statistics and Data Analysis

J. Cwik and J. Koronacki. Institute of Computer Science, Polish Academy of Sciences. to appear in. Computational Statistics and Data Analysis A Combined Adaptive-Mixtures/Plug-In Estimator of Multivariate Probability Densities 1 J. Cwik and J. Koronacki Institute of Computer Science, Polish Academy of Sciences Ordona 21, 01-237 Warsaw, Poland

More information

Shannon entropy in generalized order statistics from Pareto-type distributions

Shannon entropy in generalized order statistics from Pareto-type distributions Int. J. Nonlinear Anal. Appl. 4 (203 No., 79-9 ISSN: 2008-6822 (electronic http://www.ijnaa.semnan.ac.ir Shannon entropy in generalized order statistics from Pareto-type distributions B. Afhami a, M. Madadi

More information

The DMP Inverse for Rectangular Matrices

The DMP Inverse for Rectangular Matrices Filomat 31:19 (2017, 6015 6019 https://doi.org/10.2298/fil1719015m Published by Faculty of Sciences Mathematics, University of Niš, Serbia Available at: http://.pmf.ni.ac.rs/filomat The DMP Inverse for

More information

Jordan Journal of Mathematics and Statistics (JJMS) 7(4), 2014, pp

Jordan Journal of Mathematics and Statistics (JJMS) 7(4), 2014, pp Jordan Journal of Mathematics and Statistics (JJMS) 7(4), 2014, pp.257-271 ON RELATIONS FOR THE MOMENT GENERATING FUNCTIONS FROM THE EXTENDED TYPE II GENERALIZED LOGISTIC DISTRIBUTION BASED ON K-TH UPPER

More information

CALCULUS JIA-MING (FRANK) LIOU

CALCULUS JIA-MING (FRANK) LIOU CALCULUS JIA-MING (FRANK) LIOU Abstract. Contents. Power Series.. Polynomials and Formal Power Series.2. Radius of Convergence 2.3. Derivative and Antiderivative of Power Series 4.4. Power Series Expansion

More information

Density estimators for the convolution of discrete and continuous random variables

Density estimators for the convolution of discrete and continuous random variables Density estimators for the convolution of discrete and continuous random variables Ursula U Müller Texas A&M University Anton Schick Binghamton University Wolfgang Wefelmeyer Universität zu Köln Abstract

More information

A Conditional Approach to Modeling Multivariate Extremes

A Conditional Approach to Modeling Multivariate Extremes A Approach to ing Multivariate Extremes By Heffernan & Tawn Department of Statistics Purdue University s April 30, 2014 Outline s s Multivariate Extremes s A central aim of multivariate extremes is trying

More information

A NOTE ON RATIONAL OPERATOR MONOTONE FUNCTIONS. Masaru Nagisa. Received May 19, 2014 ; revised April 10, (Ax, x) 0 for all x C n.

A NOTE ON RATIONAL OPERATOR MONOTONE FUNCTIONS. Masaru Nagisa. Received May 19, 2014 ; revised April 10, (Ax, x) 0 for all x C n. Scientiae Mathematicae Japonicae Online, e-014, 145 15 145 A NOTE ON RATIONAL OPERATOR MONOTONE FUNCTIONS Masaru Nagisa Received May 19, 014 ; revised April 10, 014 Abstract. Let f be oeprator monotone

More information

#A29 INTEGERS 11 (2011) THE (EXPONENTIAL) BIPARTITIONAL POLYNOMIALS AND POLYNOMIAL SEQUENCES OF TRINOMIAL TYPE: PART II

#A29 INTEGERS 11 (2011) THE (EXPONENTIAL) BIPARTITIONAL POLYNOMIALS AND POLYNOMIAL SEQUENCES OF TRINOMIAL TYPE: PART II #A29 INTEGERS 11 (2011) THE (EXPONENTIAL) BIPARTITIONAL POLYNOMIALS AND POLYNOMIAL SEQUENCES OF TRINOMIAL TYPE: PART II Hacène Belbachir 1 USTHB, Faculty of Mathematics, El Alia, Bab Ezzouar, Algeria hbelbachir@usthb.dz

More information

Group-invariant solutions of nonlinear elastodynamic problems of plates and shells *

Group-invariant solutions of nonlinear elastodynamic problems of plates and shells * Group-invariant solutions of nonlinear elastodynamic problems of plates and shells * V. A. Dzhupanov, V. M. Vassilev, P. A. Dzhondzhorov Institute of mechanics, Bulgarian Academy of Sciences, Acad. G.

More information

Bivariate generalized Pareto distribution

Bivariate generalized Pareto distribution Bivariate generalized Pareto distribution in practice Eötvös Loránd University, Budapest, Hungary Minisymposium on Uncertainty Modelling 27 September 2011, CSASC 2011, Krems, Austria Outline Short summary

More information

Houston Journal of Mathematics. c 2008 University of Houston Volume 34, No. 4, 2008

Houston Journal of Mathematics. c 2008 University of Houston Volume 34, No. 4, 2008 Houston Journal of Mathematics c 2008 University of Houston Volume 34, No. 4, 2008 SHARING SET AND NORMAL FAMILIES OF ENTIRE FUNCTIONS AND THEIR DERIVATIVES FENG LÜ AND JUNFENG XU Communicated by Min Ru

More information

EXTREMAL QUANTILES OF MAXIMUMS FOR STATIONARY SEQUENCES WITH PSEUDO-STATIONARY TREND WITH APPLICATIONS IN ELECTRICITY CONSUMPTION ALEXANDR V.

EXTREMAL QUANTILES OF MAXIMUMS FOR STATIONARY SEQUENCES WITH PSEUDO-STATIONARY TREND WITH APPLICATIONS IN ELECTRICITY CONSUMPTION ALEXANDR V. MONTENEGRIN STATIONARY JOURNAL TREND WITH OF ECONOMICS, APPLICATIONS Vol. IN 9, ELECTRICITY No. 4 (December CONSUMPTION 2013), 53-63 53 EXTREMAL QUANTILES OF MAXIMUMS FOR STATIONARY SEQUENCES WITH PSEUDO-STATIONARY

More information

ONE DIMENSIONAL MARGINALS OF OPERATOR STABLE LAWS AND THEIR DOMAINS OF ATTRACTION

ONE DIMENSIONAL MARGINALS OF OPERATOR STABLE LAWS AND THEIR DOMAINS OF ATTRACTION ONE DIMENSIONAL MARGINALS OF OPERATOR STABLE LAWS AND THEIR DOMAINS OF ATTRACTION Mark M. Meerschaert Department of Mathematics University of Nevada Reno NV 89557 USA mcubed@unr.edu and Hans Peter Scheffler

More information

The Convergence Rate for the Normal Approximation of Extreme Sums

The Convergence Rate for the Normal Approximation of Extreme Sums The Convergence Rate for the Normal Approximation of Extreme Sums Yongcheng Qi University of Minnesota Duluth WCNA 2008, Orlando, July 2-9, 2008 This talk is based on a joint work with Professor Shihong

More information

An Approximate Solution for Volterra Integral Equations of the Second Kind in Space with Weight Function

An Approximate Solution for Volterra Integral Equations of the Second Kind in Space with Weight Function International Journal of Mathematical Analysis Vol. 11 17 no. 18 849-861 HIKARI Ltd www.m-hikari.com https://doi.org/1.1988/ijma.17.771 An Approximate Solution for Volterra Integral Equations of the Second

More information

Additive functionals of infinite-variance moving averages. Wei Biao Wu The University of Chicago TECHNICAL REPORT NO. 535

Additive functionals of infinite-variance moving averages. Wei Biao Wu The University of Chicago TECHNICAL REPORT NO. 535 Additive functionals of infinite-variance moving averages Wei Biao Wu The University of Chicago TECHNICAL REPORT NO. 535 Departments of Statistics The University of Chicago Chicago, Illinois 60637 June

More information

Explicit formulas for computing Bernoulli numbers of the second kind and Stirling numbers of the first kind

Explicit formulas for computing Bernoulli numbers of the second kind and Stirling numbers of the first kind Filomat 28:2 (24), 39 327 DOI.2298/FIL4239O Published by Faculty of Sciences and Mathematics, University of Niš, Serbia Available at: http://www.pmf.ni.ac.rs/filomat Explicit formulas for computing Bernoulli

More information

ON THE COMPOUND POISSON DISTRIBUTION

ON THE COMPOUND POISSON DISTRIBUTION Acta Sci. Math. Szeged) 18 1957), pp. 23 28. ON THE COMPOUND POISSON DISTRIBUTION András Préopa Budapest) Received: March 1, 1957 A probability distribution is called a compound Poisson distribution if

More information

Journal of Inequalities in Pure and Applied Mathematics

Journal of Inequalities in Pure and Applied Mathematics Journal of Inequalities in Pure and Applied Mathematics ON SIMULTANEOUS APPROXIMATION FOR CERTAIN BASKAKOV DURRMEYER TYPE OPERATORS VIJAY GUPTA, MUHAMMAD ASLAM NOOR AND MAN SINGH BENIWAL School of Applied

More information

FULL LIKELIHOOD INFERENCES IN THE COX MODEL

FULL LIKELIHOOD INFERENCES IN THE COX MODEL October 20, 2007 FULL LIKELIHOOD INFERENCES IN THE COX MODEL BY JIAN-JIAN REN 1 AND MAI ZHOU 2 University of Central Florida and University of Kentucky Abstract We use the empirical likelihood approach

More information

Seong Joo Kang. Let u be a smooth enough solution to a quasilinear hyperbolic mixed problem:

Seong Joo Kang. Let u be a smooth enough solution to a quasilinear hyperbolic mixed problem: Comm. Korean Math. Soc. 16 2001, No. 2, pp. 225 233 THE ENERGY INEQUALITY OF A QUASILINEAR HYPERBOLIC MIXED PROBLEM Seong Joo Kang Abstract. In this paper, e establish the energy inequalities for second

More information

Collatz cycles with few descents

Collatz cycles with few descents ACTA ARITHMETICA XCII.2 (2000) Collatz cycles with few descents by T. Brox (Stuttgart) 1. Introduction. Let T : Z Z be the function defined by T (x) = x/2 if x is even, T (x) = (3x + 1)/2 if x is odd.

More information

WEIBULL RENEWAL PROCESSES

WEIBULL RENEWAL PROCESSES Ann. Inst. Statist. Math. Vol. 46, No. 4, 64-648 (994) WEIBULL RENEWAL PROCESSES NIKOS YANNAROS Department of Statistics, University of Stockholm, S-06 9 Stockholm, Sweden (Received April 9, 993; revised

More information

A class of probability distributions for application to non-negative annual maxima

A class of probability distributions for application to non-negative annual maxima Hydrol. Earth Syst. Sci. Discuss., doi:.94/hess-7-98, 7 A class of probability distributions for application to non-negative annual maxima Earl Bardsley School of Science, University of Waikato, Hamilton

More information

Improved Newton s method with exact line searches to solve quadratic matrix equation

Improved Newton s method with exact line searches to solve quadratic matrix equation Journal of Computational and Applied Mathematics 222 (2008) 645 654 wwwelseviercom/locate/cam Improved Newton s method with exact line searches to solve quadratic matrix equation Jian-hui Long, Xi-yan

More information

A Type of Shannon-McMillan Approximation Theorems for Second-Order Nonhomogeneous Markov Chains Indexed by a Double Rooted Tree

A Type of Shannon-McMillan Approximation Theorems for Second-Order Nonhomogeneous Markov Chains Indexed by a Double Rooted Tree Caspian Journal of Applied Mathematics, Ecology and Economics V. 2, No, 204, July ISSN 560-4055 A Type of Shannon-McMillan Approximation Theorems for Second-Order Nonhomogeneous Markov Chains Indexed by

More information

1. Nonlinear Equations. This lecture note excerpted parts from Michael Heath and Max Gunzburger. f(x) = 0

1. Nonlinear Equations. This lecture note excerpted parts from Michael Heath and Max Gunzburger. f(x) = 0 Numerical Analysis 1 1. Nonlinear Equations This lecture note excerpted parts from Michael Heath and Max Gunzburger. Given function f, we seek value x for which where f : D R n R n is nonlinear. f(x) =

More information

PREPRINT 2005:38. Multivariate Generalized Pareto Distributions HOLGER ROOTZÉN NADER TAJVIDI

PREPRINT 2005:38. Multivariate Generalized Pareto Distributions HOLGER ROOTZÉN NADER TAJVIDI PREPRINT 2005:38 Multivariate Generalized Pareto Distributions HOLGER ROOTZÉN NADER TAJVIDI Department of Mathematical Sciences Division of Mathematical Statistics CHALMERS UNIVERSITY OF TECHNOLOGY GÖTEBORG

More information

J. Combin. Theory Ser. A 116(2009), no. 8, A NEW EXTENSION OF THE ERDŐS-HEILBRONN CONJECTURE

J. Combin. Theory Ser. A 116(2009), no. 8, A NEW EXTENSION OF THE ERDŐS-HEILBRONN CONJECTURE J. Combin. Theory Ser. A 116(2009), no. 8, 1374 1381. A NEW EXTENSION OF THE ERDŐS-HEILBRONN CONJECTURE Hao Pan and Zhi-Wei Sun Department of Mathematics, Naning University Naning 210093, People s Republic

More information

Tail properties and asymptotic distribution for maximum of LGMD

Tail properties and asymptotic distribution for maximum of LGMD Tail properties asymptotic distribution for maximum of LGMD a Jianwen Huang, b Shouquan Chen a School of Mathematics Computational Science, Zunyi Normal College, Zunyi, 56300, China b School of Mathematics

More information

On variable bandwidth kernel density estimation

On variable bandwidth kernel density estimation JSM 04 - Section on Nonparametric Statistics On variable bandwidth kernel density estimation Janet Nakarmi Hailin Sang Abstract In this paper we study the ideal variable bandwidth kernel estimator introduced

More information

Global Attractivity in a Higher Order Nonlinear Difference Equation

Global Attractivity in a Higher Order Nonlinear Difference Equation Applied Mathematics E-Notes, (00), 51-58 c ISSN 1607-510 Available free at mirror sites of http://www.math.nthu.edu.tw/ amen/ Global Attractivity in a Higher Order Nonlinear Difference Equation Xing-Xue

More information

The Goodness-of-fit Test for Gumbel Distribution: A Comparative Study

The Goodness-of-fit Test for Gumbel Distribution: A Comparative Study MATEMATIKA, 2012, Volume 28, Number 1, 35 48 c Department of Mathematics, UTM. The Goodness-of-fit Test for Gumbel Distribution: A Comparative Study 1 Nahdiya Zainal Abidin, 2 Mohd Bakri Adam and 3 Habshah

More information

THE DENOMINATORS OF POWER SUMS OF ARITHMETIC PROGRESSIONS. Bernd C. Kellner Göppert Weg 5, Göttingen, Germany

THE DENOMINATORS OF POWER SUMS OF ARITHMETIC PROGRESSIONS. Bernd C. Kellner Göppert Weg 5, Göttingen, Germany #A95 INTEGERS 18 (2018) THE DENOMINATORS OF POWER SUMS OF ARITHMETIC PROGRESSIONS Bernd C. Kellner Göppert Weg 5, 37077 Göttingen, Germany b@bernoulli.org Jonathan Sondow 209 West 97th Street, New Yor,

More information

Cell throughput analysis of the Proportional Fair scheduler in the single cell environment

Cell throughput analysis of the Proportional Fair scheduler in the single cell environment Cell throughput analysis of the Proportional Fair scheduler in the single cell environment Jin-Ghoo Choi and Seawoong Bahk IEEE Trans on Vehicular Tech, Mar 2007 *** Presented by: Anh H. Nguyen February

More information

Multivariate Pareto distributions: properties and examples

Multivariate Pareto distributions: properties and examples Multivariate Pareto distributions: properties and examples Ana Ferreira 1, Laurens de Haan 2 1 ISA UTL and CEAUL, Portugal 2 Erasmus Univ Rotterdam and CEAUL EVT2013 Vimeiro, September 8 11 Univariate

More information

Max stable processes: representations, ergodic properties and some statistical applications

Max stable processes: representations, ergodic properties and some statistical applications Max stable processes: representations, ergodic properties and some statistical applications Stilian Stoev University of Michigan, Ann Arbor Oberwolfach March 21, 2008 The focus Let X = {X t } t R be a

More information

GERHARD S I E GEL (DRESDEN)

GERHARD S I E GEL (DRESDEN) PROBABILITY AND MATHEMATICAI, STATISTICS Vol. 13, Fax. 1 (1992), pp 33-37 ON THE CLASS OF OPERATOR STABLE DISTRIBUTIONS IN A SEPARABLE BANACH SPACE GERHARD S I E GEL (DRESDEN) Abstract. This paper characterizes

More information

EXISTENCE OF POSITIVE PERIODIC SOLUTIONS OF DISCRETE MODEL FOR THE INTERACTION OF DEMAND AND SUPPLY. S. H. Saker

EXISTENCE OF POSITIVE PERIODIC SOLUTIONS OF DISCRETE MODEL FOR THE INTERACTION OF DEMAND AND SUPPLY. S. H. Saker Nonlinear Funct. Anal. & Appl. Vol. 10 No. 005 pp. 311 34 EXISTENCE OF POSITIVE PERIODIC SOLUTIONS OF DISCRETE MODEL FOR THE INTERACTION OF DEMAND AND SUPPLY S. H. Saker Abstract. In this paper we derive

More information

Characterizations of Pareto, Weibull and Power Function Distributions Based On Generalized Order Statistics

Characterizations of Pareto, Weibull and Power Function Distributions Based On Generalized Order Statistics Marquette University e-publications@marquette Mathematics, Statistics and Computer Science Faculty Research and Publications Mathematics, Statistics and Computer Science, Department of 8--26 Characterizations

More information

A REMARK ON ROBUSTNESS AND WEAK CONTINUITY OF M-ESTEVtATORS

A REMARK ON ROBUSTNESS AND WEAK CONTINUITY OF M-ESTEVtATORS J. Austral. Math. Soc. (Series A) 68 (2000), 411-418 A REMARK ON ROBUSTNESS AND WEAK CONTINUITY OF M-ESTEVtATORS BRENTON R. CLARKE (Received 11 January 1999; revised 19 November 1999) Communicated by V.

More information

Exact Asymptotics in Complete Moment Convergence for Record Times and the Associated Counting Process

Exact Asymptotics in Complete Moment Convergence for Record Times and the Associated Counting Process A^VÇÚO 33 ò 3 Ï 207 c 6 Chinese Journal of Applied Probability Statistics Jun., 207, Vol. 33, No. 3, pp. 257-266 doi: 0.3969/j.issn.00-4268.207.03.004 Exact Asymptotics in Complete Moment Convergence for

More information

arxiv: v2 [math.fa] 17 May 2016

arxiv: v2 [math.fa] 17 May 2016 ESTIMATES ON SINGULAR VALUES OF FUNCTIONS OF PERTURBED OPERATORS arxiv:1605.03931v2 [math.fa] 17 May 2016 QINBO LIU DEPARTMENT OF MATHEMATICS, MICHIGAN STATE UNIVERSITY EAST LANSING, MI 48824, USA Abstract.

More information

Journal of Inequalities in Pure and Applied Mathematics

Journal of Inequalities in Pure and Applied Mathematics Journal of Inequalities in Pure and Applied Mathematics INEQUALITIES FOR GENERAL INTEGRAL MEANS GHEORGHE TOADER AND JOZSEF SÁNDOR Department of Mathematics Technical University Cluj-Napoca, Romania. EMail:

More information

Observer design for a general class of triangular systems

Observer design for a general class of triangular systems 1st International Symposium on Mathematical Theory of Networks and Systems July 7-11, 014. Observer design for a general class of triangular systems Dimitris Boskos 1 John Tsinias Abstract The paper deals

More information

Giaccardi s Inequality for Convex-Concave Antisymmetric Functions and Applications

Giaccardi s Inequality for Convex-Concave Antisymmetric Functions and Applications Southeast Asian Bulletin of Mathematics 01) 36: 863 874 Southeast Asian Bulletin of Mathematics c SEAMS 01 Giaccardi s Inequality for Convex-Concave Antisymmetric Functions and Applications J Pečarić Abdus

More information

CHARACTERIZATIONS OF UNIFORM AND EXPONENTIAL DISTRIBUTIONS VIA MOMENTS OF THE kth RECORD VALUES RANDOMLY INDEXED

CHARACTERIZATIONS OF UNIFORM AND EXPONENTIAL DISTRIBUTIONS VIA MOMENTS OF THE kth RECORD VALUES RANDOMLY INDEXED APPLICATIONES MATHEMATICAE 24,3(1997), pp. 37 314 Z. GRUDZIEŃ and D. SZYNAL(Lublin) CHARACTERIZATIONS OF UNIFORM AND EXPONENTIAL DISTRIBUTIONS VIA MOMENTS OF THE kth RECORD VALUES RANDOMLY INDEXED Abstract.

More information

On Some Estimates of the Remainder in Taylor s Formula

On Some Estimates of the Remainder in Taylor s Formula Journal of Mathematical Analysis and Applications 263, 246 263 (2) doi:.6/jmaa.2.7622, available online at http://www.idealibrary.com on On Some Estimates of the Remainder in Taylor s Formula G. A. Anastassiou

More information

Yunhi Cho and Young-One Kim

Yunhi Cho and Young-One Kim Bull. Korean Math. Soc. 41 (2004), No. 1, pp. 27 43 ANALYTIC PROPERTIES OF THE LIMITS OF THE EVEN AND ODD HYPERPOWER SEQUENCES Yunhi Cho Young-One Kim Dedicated to the memory of the late professor Eulyong

More information

On Bayesian Inference with Conjugate Priors for Scale Mixtures of Normal Distributions

On Bayesian Inference with Conjugate Priors for Scale Mixtures of Normal Distributions Journal of Applied Probability & Statistics Vol. 5, No. 1, xxx xxx c 2010 Dixie W Publishing Corporation, U. S. A. On Bayesian Inference with Conjugate Priors for Scale Mixtures of Normal Distributions

More information

CONVOLUTION OPERATORS IN INFINITE DIMENSION

CONVOLUTION OPERATORS IN INFINITE DIMENSION PORTUGALIAE MATHEMATICA Vol. 51 Fasc. 4 1994 CONVOLUTION OPERATORS IN INFINITE DIMENSION Nguyen Van Khue and Nguyen Dinh Sang 1 Introduction Let E be a complete convex bornological vector space (denoted

More information

Chapter 2: Fundamentals of Statistics Lecture 15: Models and statistics

Chapter 2: Fundamentals of Statistics Lecture 15: Models and statistics Chapter 2: Fundamentals of Statistics Lecture 15: Models and statistics Data from one or a series of random experiments are collected. Planning experiments and collecting data (not discussed here). Analysis:

More information

A combinatorial problem related to Mahler s measure

A combinatorial problem related to Mahler s measure A combinatorial problem related to Mahler s measure W. Duke ABSTRACT. We give a generalization of a result of Myerson on the asymptotic behavior of norms of certain Gaussian periods. The proof exploits

More information