Best Linear Unbiased and Invariant Reconstructors for the Past Records
|
|
- Julius Barker
- 5 years ago
- Views:
Transcription
1 BULLETIN of the MALAYSIAN MATHEMATICAL SCIENCES SOCIETY Bull Malays Math Sci Soc (2) 37(4) (2014), Best Linear Unbiased and Invariant Reconstructors for the Past Records 1 B KHATIB AND 2 JAFAR AHMADI 1,2 Departent of Statistics, Ordered and Spatial Data Center of Excellence, Ferdowsi University of Mashhad, P O Box 1159, Mashhad 91775, Iran 1 khatib b@yahooco, 2 ahadi-j@uacir Abstract The best linear unbiased reconstructor and the best linear invariant reconstructor of the past upper records based on observed upper record values for the location scale faily of distributions are derived The results are obtained in detail for two-paraeter exponential distribution A coparison study is perfored in ters of ean squared reconstruction error and Pitan s easure of closeness criteria Finally, a real exaple is given to illustrate the proposed procedures 1 Introduction 2010 Matheatics Subject Classification: 62G30 (62N99) Keywords and phrases: Best linear unbiased estiator (BLUE), best linear invariant estiator (BLIE), Pitan closeness, location-scale faily Let {X i,i > 1} be a sequence of independent and identically distributed (iid) continuous rando variables An observation X j is called an upper record value if its value exceeds that of all previous observations, i e, X j is an upper record value if X j > X i for every i < j These type of data are of great iportance in several real-life probles involving weather, industry, econoic and sport data For ore details and applications of record values, see for exaple, Arnold et al [4] Ahsanullah and Shakil [3] obtained soe characterization results of Rayleigh distribution based on conditional expectation of record values The prediction proble of the future records based on past records have been extensively studied in both frequentist and Bayesian approaches; Awad and Raqab [6] considered the prediction proble of the future nth record value based on the first ( < n) observed record values fro one-paraeter exponential distribution Raqab [17] established different point predictors and prediction intervals for future records based on observed record values fro two-paraeter exponential distribution Raqab and Balakrishnan [18] discussed the proble of prediction of the future record values in nonparaetric settings Ahadi and Doosparast [2] investigated Bayesian estiation and prediction for soe life distributions based on upper record values MirMostafaee and Ahadi [16] obtained several point Counicated by M Ataharul Isla Received: March 1, 2012; Revised: October 2, 2012
2 1018 B Khatib and J Ahadi predictors such as the best unbiased predictor, the best invariant predictor and axiu likelihood predictor of future order statistics on the basis of observed record values fro two-paraeter exponential distribution Recently, soe works have been done on the reconstruction proble Kliczak and Rychlik [14] obtained upper bounds for the expectations of increents of past order statistics and records based on observed order statistics and records, respectively Balakrishnan et al [10] investigated the proble of reconstructing past upper records fro the known values of future records in exponential and Pareto distributions Razkhah et al [19] obtained soe point and interval reconstructors for the issing order statistics in exponential distribution Asgharzadeh et al [5] studied the reconstruction of the past failure ties based on left-censored saples for the proportional reversed hazard rate odels The ain goal of this paper is to investigate the best linear unbiased reconstructor (BLUR) and the best linear invariant reconstructor (BLIR) based on observed future record data fro two-paraeter exponential distribution We also intend to present a coparison study of coon reconstructors of past records fro exponential distribution With this in ind, we consider two criteria: ean squared reconstruction error (MSRE) and Pitan s easure of closeness (PMC) In recent years, any probles involving ordered data and Pitan closeness have been investigated Pitan closeness of records to population quantiles was explored in Ahadi and Balakrishnan [1] Soe interesting results regarding the saple edian and Pitan closeness were provided by Iliopoulous and Balakrishnan [13] Based on a Type-II right censored saple fro one-paraeter exponential distribution, Balakrishnan et al [9] copared the best linear unbiased predictor (BLUP) and the best linear invariant predictor (BLIP) of the censored order statistics in the one-saple case in ters of PMC The rest of this paper is organized as follows Section 2 contains soe preliinaries In Sections 3 and 4, the BLUR and BLIR of the past upper records fro two-paraeter exponential distribution are investigated, respectively These two reconstructors and two others where obtained by Balakrishnan et al [10] are copared in ters of MSRE and PMC criteria in Section 5 A real exaple is given in Section 6 2 Preliinaries Let R 1,,R n be the first n upper record values coing fro a X-sequence of iid continuous rando variables with cuulative distribution function (cdf) F(x; θ) and probability density function (pdf) f (x;θ), where θ Θ ay be a vector of paraeters and Θ is the paraeter space Suppose we have failed to observe the first records, naely, we have observed R = (R +1,R +2,,R n ), + 1 < n In this case, the joint pdf of R can be expressed as follows (see, for exaple, Arnold et al [4, pp 10 11]) (21) ( H(r+1 ;θ) ) f R (r;θ) = f (r +1 ;θ)! n 1 i=+1 f (r i+1 ;θ) F(r i ;θ), r +1 < < r n, where F(x;θ) = 1 F(x;θ) is the survival function of X, H(x;θ) = log F(x;θ), r = (r +1,,r n ) is the observed value of R Fro (21) and using the Markovian property of record values, it can be shown that the conditional pdf of R l given the observed data set R is just the conditional pdf of R l given R +1 and is given by f Rl R(r l r;θ) = f Rl R +1 (r l r +1 ;θ)
3 (22) Best Linear Unbiased and Invariant Reconstructors for the Past Records 1019 = ( H(rl ;θ) ) l 1( H(r+1 ;θ) H(r l ;θ) ) l B(l, l + 1) ( H(r +1 ;θ) ) f (r l ;θ) F(r l ;θ), r l < r +1, where B(,) is the coplete beta function Fro (22), the unbiased reconstructor of R l, l = 1,,, is the conditional ean of R l given R +1, ie, (23) R l = E(R l R) = E(R l R +1 ), and hence it depends only on R +1 If the paraeters in (23) are unknown, they have to be estiated in this conditional expectation A rando variable X is said to have a two-paraeter exponential distribution, denoted by Exp(µ,σ), if its cdf is (24) F(x;θ) = 1 e x µ σ, x µ, σ > 0, where θ = (µ,σ), µ and σ are the location and scale paraeters, respectively The exponential distribution is a very coonly used distribution in reliability engineering and survival analysis We refer the reader to Balakrishnan and Basu [7] for soe research based on exponential distribution In what follows, we shall study the behavior of reconstructors of the past records for Exp(µ,σ) in detail The first result for the case of Exp(µ,σ) is that the unbiased reconstructor of R l based on R is given by (25) R l (µ) = µ + l + 1 (R +1 µ), it does not depend on the scale paraeter When µ is unknown, we can substitute it with its estiator based on R = (R +1,R +2,,R n ) 3 The best linear unbiased reconstructor In this section, first we focus our attention to derive the BLUR in general for the location scale failies Let X = (X +1,,X n ) be an observed ()-diensional rando vector fro the cdf F(x; µ,σ) = F 0 ((x µ)/σ), where µ and σ are the location and scale paraeters, respectively, and F 0 does not depend on µ and σ Also, let Y be an unknown past observation fro the sae distribution Reconstructing the outcoe Y based on the observed value of X is the ain goal of this section Toward this end, let us assue that a 0 = E((Y µ)/σ),a i = E((X i µ)/σ),w i = Cov((Y µ)/σ,(x i µ)/σ) and v i, j = Cov((X i µ)/σ,(x j µ)/σ), for i, j = + 1,,n Denote by 1 the vector of 1, s, a = (a +1,,a n ) T, w T = (w +1,,w n ) and V = (v i, j ),i, j = + 1,,n To obtain the BLUR of Y based on X, we will have to iniize the variance of (Y cx T ) subject to the condition of unbiasedness where c = (c +1,,c n ) is an arbitrary vector of real values Then, following along the sae lines as in the prediction proble, the BLUR of Y is given by (see, Goldberger [12] in the context of prediction) (31) Ŷ = ˆµ(X) + ˆσ(X)a 0 + w T V 1 [X T ˆµ(X)1 ˆσ(X)a], where ˆµ(X) and ˆσ(X) are the BLUEs of µ and σ on the basis of X, respectively, and V 1 is the inverse atrix of V Now, assue that the observed data are the upper record values fro the two-paraeter exponential distribution Using (31), the BLUR of R l based on R is given by (32) ˆR l = ˆµ + l ˆσ + ω T Σ 1 [R T ˆµ1 ˆσα],
4 1020 B Khatib and J Ahadi where ˆµ and ˆσ are the BLUEs of µ and σ based on R, respectively, α is the vector of eans of the corresponding record values taken fro Exp(0,1), σ 2 Σ is the covariance atrix of R and ω T = 1/σ 2 (Cov(R +1,R l ),Cov(R +2,R l ),,Cov(R n,r l )) By doing soe algebraic calculations it can be shown that ω T = (l,l,,l) and (33) ω T Σ 1 l = ( + 1,0,0,,0) Substituting (33) into (32), we get (34) ˆR l = ˆµ + l ˆσ Furtherore, the BLUEs of µ and σ on the basis of R are given by (35) ˆµ = nr +1 ( + 1)R n n ( + 1) and ˆσ = R n R +1 n ( + 1), respectively (see, Arnold et al [4]) Therefore, by substituting (35) in (34), it is deduced that (36) ˆR l = (n l)r +1 (( + 1) l)r n n ( + 1) is the BLUR of R l on the basis of the data set R The MSRE of ˆR l is coputed to be (37) MSRE( ˆR l ) = E[ ˆR l R l ] 2 = σ 2 (n l)(( + 1) l) n ( + 1) It is easy to show that MSRE( ˆR l ) is increasing in and decreasing with respect to n and l, if the other arguents are fixed Moreover, ˆR l and MSRE( ˆR l ) converge to R +1 ˆσ and σ 2 ()/(n ( + 1)), respectively, when l tends to Reark 31 It ay be noted that by substituting the BLUE of µ into the unbiased reconstructor of R l in (25), the sae result will be obtained for ˆR l as in (36), ie, R l ( ˆµ) = ˆR l 4 The best linear invariant reconstructor Sae as in previous section, first we present the results for BLIR in general for location scale failies Let X = (X +1,,X n ) and Y be as defined in Section 3 Then, follow the results of Mann [15], in the context of prediction, the BLIR of Y based on the data set X can be derived as (41) Ỹ = Ŷ ( c c 2 ) ˆσ(X), where Ŷ is the BLUR of Y, c 2 = σ 2 Var( ˆσ(X)) and c 1 = 1/σ 2 Cov( ˆσ(X),(1 w T V 1 1) ˆµ(X) + (a 0 w T V 1 a) ˆσ(X)), with V, w, a, ˆµ(X), ˆσ(X) and a 0 being as defined in Section 3 When the upper record statistics R = (R +1,R +2,,R n ) fro the Exp(µ,σ) distribution are observed, using (41), the BLIR of R l on the basis of R is as follows (42) R l = ˆR l ( V 1 1 +V 2 ) ˆσ,
5 Best Linear Unbiased and Invariant Reconstructors for the Past Records 1021 where V 2 = σ 2 Var( ˆσ) and V 1 = σ 2[ Cov( ˆσ,(1 ω T Σ 1 1) ˆµ +(l ω T Σ 1 α) ˆσ) ] in which ω,σ 1 and α are as defined in Section 3, ˆµ and ˆσ are the BLUEs of µ and σ, respectively, and ˆR l is the BLUR of R l By perforing soe algebraic calculations, we obtain ( ) n + 1 l R+1 ( + 1 l ) R n (43) R l = Moreover, it can be shown that the MSRE of R l is given by ( )( ) n + 1 l + 1 l (44) MSRE( R l ) = E[ R l R l ] 2 = σ 2, which is increasing with respect to and decreasing with respect to n and l, if the other arguents are fixed Also, R l and MSRE( R l ) converge to R +1 σ and σ 2 (1+1/(n )), respectively, when l tends to Reark 41 It ay be noted that the BLIEs of µ and σ for a two-paraeter exponential distribution are given by (see, Arnold et al [4]) (45) µ = (n + 1)R +1 ( + 1)R n and σ = R n R +1, respectively By plugging the BLIE of µ into (25), we arrive at the sae result for R l as given in (43), ie, R l ( µ) = R l 5 Coparison results In Sections 3 and 4, we obtained two reconstructors (BLUR and BLIR) for R l based on R fro a two-paraeter exponential distribution Balakrishnan et al [10] also considered the sae plan and obtained two other reconstructors for R l naely axiu likelihood reconstructor (MLR), denoted by ˆR l,m, and conditional edian reconstructor (CMR), denoted by ˆR l,c For a coparison study, let us restate their results here The MLR of R l is (51) ˆR l,m = (n l + 1)R +1 ( l)r n + 1 with (52) MSRE( ˆR l,m ) = σ 2 { (n + 1 l)( + 1 l) + 1 The CMR of R l is + } (n l)(n + 1 l) ( + 1) 2 (53) ˆR l,c = (R n R +1 )M l, + nr +1 R n with ( ) (1 MSRE( ˆR l,c ) = σ {( 2 Ml, ) l)( + 2 l) + ( 1) ( ) } (1 Ml, ) (54) () 2 ( + 1 l)( 1), where M l, stands for the edian of beta distribution with paraeters l and l + 1 In this section, we intend to copare the four reconstructors based on MSRE and PMC criteria
6 1022 B Khatib and J Ahadi 51 Coparison based on MSRE Using (37) and (44), we have (55) MSRE( R l ) = MSRE( ˆR l ) σ 2 ( ( + 1) l ) 2 (n ( + 1))() Fro (55), it is obvious that BLIR is better than BLUR in the sense of MSRE Moreover, fro (44) and (52), we have ( ) n + 1 l 2 (56) MSRE( R l ) = MSRE( ˆR l,m ) σ 2 ( ), which iplies that BLIR is better than MLR Furtherore, using (37) and (52), we get (57) ( ) MSRE( ˆR l ) = MSRE( ˆR l,m ) σ 2 n l ( + 1) 2 ( 1) () 2 (n + 2(l 1)) Using (57) it is deduced that BLUR is better than MLR when (n ) 2 > (n+ 2(l 1)) In order to coplete the coparison study, we have coputed the nuerical values of MSREs of four reconstructors for n = 10 and all selected values of l and The results are presented in Table 1 Table 1 Nuerical values of MSREs for n = ,2,3, and 4 stand for the σ 2 MSRE of ˆR l,m, ˆR l,c, ˆR l and R l, respectively For n = 10 fro Table 1, by epirical evidence, it is observed that in the sense of MSRE, BLIR is ore accurate reconstructor than others Furtherore, in the ost cases the MSRE of ˆR l,m is greater than others
7 Best Linear Unbiased and Invariant Reconstructors for the Past Records Coparison based on PMC Here, we use the PMC criterion to copare the reconstructors ˆR l, R l, ˆR l,m and ˆR l,c The easure is based on the probabilities of the relative closeness of copeting estiators to an unknown paraeter First, we recall two definitions If T 1 and T 2 are two estiators of a coon paraeter θ, PMC of T 1 relative to T 2 is defined by (58) PMC(T 1,T 2 θ) = P( T 1 θ < T 2 θ ), θ Ω, where Ω is the paraeter space The estiator T 1 is called Pitan closer (with respect to θ) than T 2 if and only if PMC(T 1,T 2 θ) 1/2, θ Ω, with strict inequality holding for at least one θ Based on a Type-II right censored saple (X 1:n,X 2:n,,X r:n ) fro Exp(0,σ) distribution, Balakrishnan et al [8] copared BLUE and BLIE of σ in ters of PMC criterion Also, Balakrishnan et al [9] copared BLUP and BLIP of future order statistics under Type-II censoring for Exp(0,σ) distribution in ters of PMC criterion Now, we copare each two reconstructors of R l based on PMC criterion (i) PMC of ˆR l relative to R l (BLUR and BLIR): Fro (36), (43) and (58), we have π ˆR l, R l (l,;n) = P( ˆR l R l < R l R l ) = P ( R l 1 (R n R +1 ) R l < R l ) (R n R +1 ) R l ( + 1 l = P ( 1) 2 (R 2 n R +1 ) 1 (R +1 R l ) < + 1 l () 2 (R n R +1 ) 2 ) (R +1 R l ) ( ( Rn R = P < R +1 R l ( + 1 l) ) 1 ) It can be shown that (R n R +1 )/(R +1 R l ) has the F-distribution with 2( 1) and 2( + 1 l) degrees of freedo So, we express π ˆR l, R (l,;n) as l ( ) 2()( 1) (59) π ˆR l, R (l,;n) = F l 2(n 1),2(+1 l), ( + 1 l)(2n 2 1) where F a,b (x) stands for the cdf of F-distribution with a and b degrees of freedo at x (ii) PMC of ˆR l relative to ˆR l,m (BLUR and MLR): Fro (36), (51) and (58), we have π ˆR l, ˆR l,m (l,;n) = P( ˆR l R l < ˆR l,m R l ) = P ( R l 1 (R n R +1 ) R l < R +1 l + 1 ) (R n R +1 ) R l
8 1024 B Khatib and J Ahadi (510) ( + 1 l = F 2(n 1),2(+1 l) (2 1 + l ) 1 ) + 1 (iii) PMC of ˆR l relative to ˆR l,c (BLUR and CMR): Fro (36), (53) and (58), we have (511) π ˆR l, ˆR l,c (l,;n) = P( ˆR l R l < ˆR l,c R l ) { = P ( R l 1 (R n R +1 ) R l < R +1 (1 M } l,) (R n R +1 ) R l ( + 1 l = F 2(n 1),2(+1 l) (2 1 + (1 M ) 1 ) l,) (iv) PMC of R l relative to ˆR l,m (BLIR and MLR): Siilarly, we have (512) π R l, ˆR l,m (l,;n) = P( R l R l < ˆR l,m R l ) = P ( R l (R n R +1 ) R l < R +1 l + 1 ) (R n R +1 ) R l = F 2(n 1),2(+1 l) (( + 1 l + l + 1 (v) PMC of R l relative to ˆR l,c (BLIR and CMR): ( + 1 l (513) π R l, ˆR (l,;n) = F l,c 2(n 1),2(+1 l) (2 (vi) PMC of ˆR l,m relative to ˆR l,c (MLR and CMR): ( l (514) π ˆR l,c, ˆR l,m (l,;n) = F 2(n 1),2(+1 l) (2 ) 1 ) + (1 M ) 1 ) l,) (1 M l,) ) 1 ) Fro the above results, one can exaine the perforance of any two reconstructors in ters of PMC criterion by coparing the values of edian of the F-distribution with 2(n 1) and 2(+1 l) degrees of freedo and corresponding values in (59) (514), but, nuerical coputations are needed We have coputed the PMC probabilities in (59) (514) for n = 5,10,15,20 and all selected values of l and The results for n = 10 are presented in Tables 2 7, respectively, which exaine how uch a reconstructor is better than the other The nuerical results for n = 5,15 and 20 are available with the authors, and will be sent on request to interested readers
9 Best Linear Unbiased and Invariant Reconstructors for the Past Records 1025 Table 2 Values of π ˆR l, R l (l,;n) for n = Table 3 Values of π ˆR l, ˆR l,m (l,;n) for n = Table 4 Values of π ˆR l, ˆR l,c (l,;n) for n = Table 5 Values of π R l, ˆR l,m (l,;n) for n =
10 1026 B Khatib and J Ahadi Table 6 Values of π R l, ˆR l,c (l,;n) for n = Table 7 Values of π ˆR l,c, ˆR l,m (l,;n) for n = Fro our nuerical results and by epirical evidence, we observe that: (1) The BLIR (CMR) is Pitan-closer than BLUR (BLUR) for the following situations: (i) when n = 10,20, = n/2 + a and l = 1,2,,2a + 1,(a = 0,1,,n/2 2); (ii) when n = 5,15, = (n + 1)/2 + a and l = 1,2,,2a + 2,(a = 0,1,,(n + 1)/2 3); otherwise, the BLUR (BLUR) is Pitan-closer than the BLIR (CMR) specially for l = [See for exaple Table 2 (Table 4) when n = 10] (2) The MLR (CMR) is Pitan closer than BLUR (BLIR) in the following situations: (i) when n = 10,20, = n/2 + a and l = 1,2,,2a,(a = 1,,n/2 2); (ii) when n = 5,15, = (n + 1)/2 + a and l = 1,2,,2a + 1,(a = 0,1,,(n + 1)/2 3); otherwise, the BLUR (BLIR) is Pitan-closer than the MLR (CMR) specially for l = [See for exaple Table 3 (Table 6) when n = 10] (3) The MLR (MLR) is Pitan-closer than BLIR (CMR) for the following situations: (i) when n = 10,20, = n/2 + a and l = 1,2,,2a 1,(a = 1,,n/2 2); (ii) when n = 5,15, = (n+1)/2+a and l = 1,2,,2a,(a = 1,,(n+1)/2 3); otherwise, the BLIR (CMR) is Pitan-closer than the MLR (MLR) specially for l = [See for exaple Table 5 (Table 7) when n = 10]
11 Best Linear Unbiased and Invariant Reconstructors for the Past Records A real exaple To illustrate the perforance of the proposed reconstructors, we use a real data set concerning the ties (in inutes) between 48 consecutive telephone calls to a copany s switchboard which is presented by Castillo et al [11] where the authors found that the exponential distribution gave an adequate fit Table 8 contains the corresponding data Table 8 Ties (in inutes) between 48 consecutive calls Fro the data of Table 8, six upper records extracted which are 134, 168, 186, 220, 320 and 325 Assuing that only R +1,,R 6 have been observed, we show that how one can reconstruct the issing record values and copare the results with the exact values With this in ind, we copute different point reconstructors presented in the paper for the lost record value R l (1 l ) Using (36), (43), (51) and (53) the values of the BLUR, BLIR, MLR and CMR are obtained, respectively The results are presented in Table 9 for = 3 and 4 and 1 l Table 9 The nuerical values of point reconstructors l Exact value ˆR l R l ˆR l,m ˆR l,c For = 3, the observed records are R 4 = 220,R 5 = 320 and R 6 = 325 Based on these observations we have reconstructed the first three records Fro Table 9, it is observed that for = 3 the point reconstructors are reasonably close to the exact values When = 4, it is assued that two records R 5 = 320 and R 6 = 325 are observed In this case, the point reconstructors are not close to the exact values but not too far It should be entiond that one exaple does not tell us uch ore Acknowledgeent The authors would like to thank the Editor in Chief and the referees for their careful reading and valuable suggestions that iproved the original version of the anuscript References [1] J Ahadi and N Balakrishnan, Pitan closeness of record values to population quantiles, Statist Probab Lett 79 (2009), no 19,
12 1028 B Khatib and J Ahadi [2] J Ahadi and M Doostparast, Bayesian estiation and prediction for soe life distributions based on record values, Statist Papers 47 (2006), no 3, [3] M Ahsanullah and M Shakil, Characterizations of Rayleigh distribution based on order statistics and record values, Bull Malays Math Sci Soc (2) 36 (2013), no 3, [4] B C Arnold, N Balakrishnan and H N Nagaraja, Records, Wiley Series in Probability and Statistics: Probability and Statistics, Wiley, New York, 1998 [5] A Asgharzadeh, J Ahadi, Z M Ganji and R Valiollahi, Reconstruction of the past failure ties for the proportional reversed hazard rate odel, J Stat Coput Siul 82 (2012), no 3, [6] A M Awad and M Z Raqab, Prediction intervals for the future record values fro exponential distribution: coparative study, J Statist Coput Siulation 65 (2000), no 4, [7] N Balakrishnan and A P Basu, The Exponential Distribution, Gordon and Breach Publishers, Asterda, 1995 [8] N Balakrishnan, K F Davies, J P Keating and R L Mason, Pitan closeness, onotonicity and consistency of best linear unbiased and invariant estiators for exponential distribution under type II censoring, J Stat Coput Siul 81 (2011), no 8, [9] N Balakrishnan, K F Davies, J P Keating and R L Mason, Pitan closeness of best linear unbiased and invariant predictors for exponential distribution in one- and two-saple situations, Counications in Statistics: Theory and Methods 41 (2012), 1 15 [10] N Balakrishnan, M Doostparast and J Ahadi, Reconstruction of past records, Metrika 70 (2009), no 1, [11] E Castillo, A S Hadi, N Balakrishnan and J M Sarabia, Extree Value and Related Models with Applications in Engineering and Science, Wiley Series in Probability and Statistics, Wiley-Interscience, Hoboken, NJ, 2005 [12] A S Goldberger, Best linear unbiased prediction in the generalized linear regression odel, J Aer Statist Assoc 57 (1962), [13] G Iliopoulos and N Balakrishnan, An odd property of saple edian fro odd saple sizes, Stat Methodol 7 (2010), no 6, [14] M Kliczak and T Rychlik, Reconstruction of previous failure ties and records, Metrika 61 (2005), no 3, [15] N R Mann, Optiu estiators for linear functions of location and scale paraeters, Ann Math Statist 40 (1969), no 6, [16] S M T K MirMostafaee and J Ahadi, Point prediction of future order statistics fro an exponential distribution, Statist Probab Lett 81 (2011), no 3, [17] M Z Raqab, Exponential distribution records: different ethods of prediction, in Recent Developents in Ordered Rando Variables, , Nova Sci Publ, New York, 2007 [18] M Z Raqab and N Balakrishnan, Prediction intervals for future records, Statist Probab Lett 78 (2008), no 13, [19] M Razkhah, B Khatib and J Ahadi, Reconstruction of order statistics in exponential distribution, J Iran Stat Soc (JIRSS) 9 (2010), no 1, 21 40
Best linear unbiased and invariant reconstructors for the past records
Best linear unbiased and invariant reconstructors for the past records B. Khatib and Jafar Ahmadi Department of Statistics, Ordered and Spatial Data Center of Excellence, Ferdowsi University of Mashhad,
More informationMachine Learning Basics: Estimators, Bias and Variance
Machine Learning Basics: Estiators, Bias and Variance Sargur N. srihari@cedar.buffalo.edu This is part of lecture slides on Deep Learning: http://www.cedar.buffalo.edu/~srihari/cse676 1 Topics in Basics
More informationNon-Parametric Non-Line-of-Sight Identification 1
Non-Paraetric Non-Line-of-Sight Identification Sinan Gezici, Hisashi Kobayashi and H. Vincent Poor Departent of Electrical Engineering School of Engineering and Applied Science Princeton University, Princeton,
More informationAN OPTIMAL SHRINKAGE FACTOR IN PREDICTION OF ORDERED RANDOM EFFECTS
Statistica Sinica 6 016, 1709-178 doi:http://dx.doi.org/10.5705/ss.0014.0034 AN OPTIMAL SHRINKAGE FACTOR IN PREDICTION OF ORDERED RANDOM EFFECTS Nilabja Guha 1, Anindya Roy, Yaakov Malinovsky and Gauri
More informationKeywords: Estimator, Bias, Mean-squared error, normality, generalized Pareto distribution
Testing approxiate norality of an estiator using the estiated MSE and bias with an application to the shape paraeter of the generalized Pareto distribution J. Martin van Zyl Abstract In this work the norality
More informationEstimation of the Population Mean Based on Extremes Ranked Set Sampling
Aerican Journal of Matheatics Statistics 05, 5(: 3-3 DOI: 0.593/j.ajs.05050.05 Estiation of the Population Mean Based on Extrees Ranked Set Sapling B. S. Biradar,*, Santosha C. D. Departent of Studies
More informationTEST OF HOMOGENEITY OF PARALLEL SAMPLES FROM LOGNORMAL POPULATIONS WITH UNEQUAL VARIANCES
TEST OF HOMOGENEITY OF PARALLEL SAMPLES FROM LOGNORMAL POPULATIONS WITH UNEQUAL VARIANCES S. E. Ahed, R. J. Tokins and A. I. Volodin Departent of Matheatics and Statistics University of Regina Regina,
More informationReconstruction of Order Statistics in Exponential Distribution
JIRSS 200) Vol. 9, No., pp 2-40 Reconstruction of Order Statistics in Exponential Distribution M. Razmkhah, B. Khatib, Jafar Ahmadi Department of Statistics, Ordered and Spatial Data Center of Excellence,
More informationBlock designs and statistics
Bloc designs and statistics Notes for Math 447 May 3, 2011 The ain paraeters of a bloc design are nuber of varieties v, bloc size, nuber of blocs b. A design is built on a set of v eleents. Each eleent
More informationOn the Maximum Likelihood Estimation of Weibull Distribution with Lifetime Data of Hard Disk Drives
314 Int'l Conf. Par. and Dist. Proc. Tech. and Appl. PDPTA'17 On the Maxiu Likelihood Estiation of Weibull Distribution with Lifetie Data of Hard Disk Drives Daiki Koizui Departent of Inforation and Manageent
More informationMoments of the product and ratio of two correlated chi-square variables
Stat Papers 009 50:581 59 DOI 10.1007/s0036-007-0105-0 REGULAR ARTICLE Moents of the product and ratio of two correlated chi-square variables Anwar H. Joarder Received: June 006 / Revised: 8 October 007
More informationEstimating Parameters for a Gaussian pdf
Pattern Recognition and achine Learning Jaes L. Crowley ENSIAG 3 IS First Seester 00/0 Lesson 5 7 Noveber 00 Contents Estiating Paraeters for a Gaussian pdf Notation... The Pattern Recognition Proble...3
More informationDEPARTMENT OF ECONOMETRICS AND BUSINESS STATISTICS
ISSN 1440-771X AUSTRALIA DEPARTMENT OF ECONOMETRICS AND BUSINESS STATISTICS An Iproved Method for Bandwidth Selection When Estiating ROC Curves Peter G Hall and Rob J Hyndan Working Paper 11/00 An iproved
More informationProbability Distributions
Probability Distributions In Chapter, we ephasized the central role played by probability theory in the solution of pattern recognition probles. We turn now to an exploration of soe particular exaples
More informationThe proofs of Theorem 1-3 are along the lines of Wied and Galeano (2013).
A Appendix: Proofs The proofs of Theore 1-3 are along the lines of Wied and Galeano (2013) Proof of Theore 1 Let D[d 1, d 2 ] be the space of càdlàg functions on the interval [d 1, d 2 ] equipped with
More informationBiostatistics Department Technical Report
Biostatistics Departent Technical Report BST006-00 Estiation of Prevalence by Pool Screening With Equal Sized Pools and a egative Binoial Sapling Model Charles R. Katholi, Ph.D. Eeritus Professor Departent
More informationInferences using type-ii progressively censored data with binomial removals
Arab. J. Math. (215 4:127 139 DOI 1.17/s465-15-127-8 Arabian Journal of Matheatics Ahed A. Solian Ahed H. Abd Ellah Nasser A. Abou-Elheggag Rashad M. El-Sagheer Inferences using type-ii progressively censored
More informationThe Distribution of the Covariance Matrix for a Subset of Elliptical Distributions with Extension to Two Kurtosis Parameters
journal of ultivariate analysis 58, 96106 (1996) article no. 0041 The Distribution of the Covariance Matrix for a Subset of Elliptical Distributions with Extension to Two Kurtosis Paraeters H. S. Steyn
More informationAN EFFICIENT CLASS OF CHAIN ESTIMATORS OF POPULATION VARIANCE UNDER SUB-SAMPLING SCHEME
J. Japan Statist. Soc. Vol. 35 No. 005 73 86 AN EFFICIENT CLASS OF CHAIN ESTIMATORS OF POPULATION VARIANCE UNDER SUB-SAMPLING SCHEME H. S. Jhajj*, M. K. Shara* and Lovleen Kuar Grover** For estiating the
More informationSymmetrization and Rademacher Averages
Stat 928: Statistical Learning Theory Lecture: Syetrization and Radeacher Averages Instructor: Sha Kakade Radeacher Averages Recall that we are interested in bounding the difference between epirical and
More informationW-BASED VS LATENT VARIABLES SPATIAL AUTOREGRESSIVE MODELS: EVIDENCE FROM MONTE CARLO SIMULATIONS
W-BASED VS LATENT VARIABLES SPATIAL AUTOREGRESSIVE MODELS: EVIDENCE FROM MONTE CARLO SIMULATIONS. Introduction When it coes to applying econoetric odels to analyze georeferenced data, researchers are well
More informationClassical and Bayesian Inference for an Extension of the Exponential Distribution under Progressive Type-II Censored Data with Binomial Removals
J. Stat. Appl. Pro. Lett. 1, No. 3, 75-86 (2014) 75 Journal of Statistics Applications & Probability Letters An International Journal http://dx.doi.org/10.12785/jsapl/010304 Classical and Bayesian Inference
More informationSPECTRUM sensing is a core concept of cognitive radio
World Acadey of Science, Engineering and Technology International Journal of Electronics and Counication Engineering Vol:6, o:2, 202 Efficient Detection Using Sequential Probability Ratio Test in Mobile
More informationModel Fitting. CURM Background Material, Fall 2014 Dr. Doreen De Leon
Model Fitting CURM Background Material, Fall 014 Dr. Doreen De Leon 1 Introduction Given a set of data points, we often want to fit a selected odel or type to the data (e.g., we suspect an exponential
More informationModified Systematic Sampling in the Presence of Linear Trend
Modified Systeatic Sapling in the Presence of Linear Trend Zaheen Khan, and Javid Shabbir Keywords: Abstract A new systeatic sapling design called Modified Systeatic Sapling (MSS, proposed by ] is ore
More informationE0 370 Statistical Learning Theory Lecture 6 (Aug 30, 2011) Margin Analysis
E0 370 tatistical Learning Theory Lecture 6 (Aug 30, 20) Margin Analysis Lecturer: hivani Agarwal cribe: Narasihan R Introduction In the last few lectures we have seen how to obtain high confidence bounds
More informationTesting equality of variances for multiple univariate normal populations
University of Wollongong Research Online Centre for Statistical & Survey Methodology Working Paper Series Faculty of Engineering and Inforation Sciences 0 esting equality of variances for ultiple univariate
More informationDERIVING PROPER UNIFORM PRIORS FOR REGRESSION COEFFICIENTS
DERIVING PROPER UNIFORM PRIORS FOR REGRESSION COEFFICIENTS N. van Erp and P. van Gelder Structural Hydraulic and Probabilistic Design, TU Delft Delft, The Netherlands Abstract. In probles of odel coparison
More informationBayes Estimation of the Logistic Distribution Parameters Based on Progressive Sampling
Appl. Math. Inf. Sci. 0, No. 6, 93-30 06) 93 Applied Matheatics & Inforation Sciences An International Journal http://dx.doi.org/0.8576/ais/0063 Bayes Estiation of the Logistic Distribution Paraeters Based
More informationare equal to zero, where, q = p 1. For each gene j, the pairwise null and alternative hypotheses are,
Page of 8 Suppleentary Materials: A ultiple testing procedure for ulti-diensional pairwise coparisons with application to gene expression studies Anjana Grandhi, Wenge Guo, Shyaal D. Peddada S Notations
More informationExperimental Design For Model Discrimination And Precise Parameter Estimation In WDS Analysis
City University of New York (CUNY) CUNY Acadeic Works International Conference on Hydroinforatics 8-1-2014 Experiental Design For Model Discriination And Precise Paraeter Estiation In WDS Analysis Giovanna
More informationProc. of the IEEE/OES Seventh Working Conference on Current Measurement Technology UNCERTAINTIES IN SEASONDE CURRENT VELOCITIES
Proc. of the IEEE/OES Seventh Working Conference on Current Measureent Technology UNCERTAINTIES IN SEASONDE CURRENT VELOCITIES Belinda Lipa Codar Ocean Sensors 15 La Sandra Way, Portola Valley, CA 98 blipa@pogo.co
More informationA Simple Regression Problem
A Siple Regression Proble R. M. Castro March 23, 2 In this brief note a siple regression proble will be introduced, illustrating clearly the bias-variance tradeoff. Let Y i f(x i ) + W i, i,..., n, where
More informationFeature Extraction Techniques
Feature Extraction Techniques Unsupervised Learning II Feature Extraction Unsupervised ethods can also be used to find features which can be useful for categorization. There are unsupervised ethods that
More information3.3 Variational Characterization of Singular Values
3.3. Variational Characterization of Singular Values 61 3.3 Variational Characterization of Singular Values Since the singular values are square roots of the eigenvalues of the Heritian atrices A A and
More informationSoft Computing Techniques Help Assign Weights to Different Factors in Vulnerability Analysis
Soft Coputing Techniques Help Assign Weights to Different Factors in Vulnerability Analysis Beverly Rivera 1,2, Irbis Gallegos 1, and Vladik Kreinovich 2 1 Regional Cyber and Energy Security Center RCES
More informationA note on the multiplication of sparse matrices
Cent. Eur. J. Cop. Sci. 41) 2014 1-11 DOI: 10.2478/s13537-014-0201-x Central European Journal of Coputer Science A note on the ultiplication of sparse atrices Research Article Keivan Borna 12, Sohrab Aboozarkhani
More informationLimited Failure Censored Life Test Sampling Plan in Burr Type X Distribution
Journal of Modern Applied Statistical Methods Volue 15 Issue 2 Article 27 11-1-2016 Liited Failure Censored Life Test Sapling Plan in Burr Type X Distribution R. R. L. Kanta Acharya Nagarjuna University,
More informationLecture October 23. Scribes: Ruixin Qiang and Alana Shine
CSCI699: Topics in Learning and Gae Theory Lecture October 23 Lecturer: Ilias Scribes: Ruixin Qiang and Alana Shine Today s topic is auction with saples. 1 Introduction to auctions Definition 1. In a single
More informationAnalyzing Simulation Results
Analyzing Siulation Results Dr. John Mellor-Cruey Departent of Coputer Science Rice University johnc@cs.rice.edu COMP 528 Lecture 20 31 March 2005 Topics for Today Model verification Model validation Transient
More informationPolygonal Designs: Existence and Construction
Polygonal Designs: Existence and Construction John Hegean Departent of Matheatics, Stanford University, Stanford, CA 9405 Jeff Langford Departent of Matheatics, Drake University, Des Moines, IA 5011 G
More informationLost-Sales Problems with Stochastic Lead Times: Convexity Results for Base-Stock Policies
OPERATIONS RESEARCH Vol. 52, No. 5, Septeber October 2004, pp. 795 803 issn 0030-364X eissn 1526-5463 04 5205 0795 infors doi 10.1287/opre.1040.0130 2004 INFORMS TECHNICAL NOTE Lost-Sales Probles with
More informationPattern Recognition and Machine Learning. Learning and Evaluation for Pattern Recognition
Pattern Recognition and Machine Learning Jaes L. Crowley ENSIMAG 3 - MMIS Fall Seester 2017 Lesson 1 4 October 2017 Outline Learning and Evaluation for Pattern Recognition Notation...2 1. The Pattern Recognition
More informationMeta-Analytic Interval Estimation for Bivariate Correlations
Psychological Methods 2008, Vol. 13, No. 3, 173 181 Copyright 2008 by the Aerican Psychological Association 1082-989X/08/$12.00 DOI: 10.1037/a0012868 Meta-Analytic Interval Estiation for Bivariate Correlations
More informationESTIMATING AND FORMING CONFIDENCE INTERVALS FOR EXTREMA OF RANDOM POLYNOMIALS. A Thesis. Presented to. The Faculty of the Department of Mathematics
ESTIMATING AND FORMING CONFIDENCE INTERVALS FOR EXTREMA OF RANDOM POLYNOMIALS A Thesis Presented to The Faculty of the Departent of Matheatics San Jose State University In Partial Fulfillent of the Requireents
More information13.2 Fully Polynomial Randomized Approximation Scheme for Permanent of Random 0-1 Matrices
CS71 Randoness & Coputation Spring 018 Instructor: Alistair Sinclair Lecture 13: February 7 Disclaier: These notes have not been subjected to the usual scrutiny accorded to foral publications. They ay
More informationBootstrapping Dependent Data
Bootstrapping Dependent Data One of the key issues confronting bootstrap resapling approxiations is how to deal with dependent data. Consider a sequence fx t g n t= of dependent rando variables. Clearly
More informationSupport Vector Machine Classification of Uncertain and Imbalanced data using Robust Optimization
Recent Researches in Coputer Science Support Vector Machine Classification of Uncertain and Ibalanced data using Robust Optiization RAGHAV PAT, THEODORE B. TRAFALIS, KASH BARKER School of Industrial Engineering
More informationTABLE FOR UPPER PERCENTAGE POINTS OF THE LARGEST ROOT OF A DETERMINANTAL EQUATION WITH FIVE ROOTS. By William W. Chen
TABLE FOR UPPER PERCENTAGE POINTS OF THE LARGEST ROOT OF A DETERMINANTAL EQUATION WITH FIVE ROOTS By Willia W. Chen The distribution of the non-null characteristic roots of a atri derived fro saple observations
More information1 Generalization bounds based on Rademacher complexity
COS 5: Theoretical Machine Learning Lecturer: Rob Schapire Lecture #0 Scribe: Suqi Liu March 07, 08 Last tie we started proving this very general result about how quickly the epirical average converges
More informationANALYSIS OF HALL-EFFECT THRUSTERS AND ION ENGINES FOR EARTH-TO-MOON TRANSFER
IEPC 003-0034 ANALYSIS OF HALL-EFFECT THRUSTERS AND ION ENGINES FOR EARTH-TO-MOON TRANSFER A. Bober, M. Guelan Asher Space Research Institute, Technion-Israel Institute of Technology, 3000 Haifa, Israel
More informationResearch Article On the Isolated Vertices and Connectivity in Random Intersection Graphs
International Cobinatorics Volue 2011, Article ID 872703, 9 pages doi:10.1155/2011/872703 Research Article On the Isolated Vertices and Connectivity in Rando Intersection Graphs Yilun Shang Institute for
More informationCS Lecture 13. More Maximum Likelihood
CS 6347 Lecture 13 More Maxiu Likelihood Recap Last tie: Introduction to axiu likelihood estiation MLE for Bayesian networks Optial CPTs correspond to epirical counts Today: MLE for CRFs 2 Maxiu Likelihood
More informationOBJECTIVES INTRODUCTION
M7 Chapter 3 Section 1 OBJECTIVES Suarize data using easures of central tendency, such as the ean, edian, ode, and idrange. Describe data using the easures of variation, such as the range, variance, and
More informationSpine Fin Efficiency A Three Sided Pyramidal Fin of Equilateral Triangular Cross-Sectional Area
Proceedings of the 006 WSEAS/IASME International Conference on Heat and Mass Transfer, Miai, Florida, USA, January 18-0, 006 (pp13-18) Spine Fin Efficiency A Three Sided Pyraidal Fin of Equilateral Triangular
More informationNote on generating all subsets of a finite set with disjoint unions
Note on generating all subsets of a finite set with disjoint unions David Ellis e-ail: dce27@ca.ac.uk Subitted: Dec 2, 2008; Accepted: May 12, 2009; Published: May 20, 2009 Matheatics Subject Classification:
More informationEstimation of the Mean of the Exponential Distribution Using Maximum Ranked Set Sampling with Unequal Samples
Open Journal of Statistics, 4, 4, 64-649 Published Online Septeber 4 in SciRes http//wwwscirporg/ournal/os http//ddoiorg/436/os4486 Estiation of the Mean of the Eponential Distribution Using Maiu Ranked
More informationA Note on the Applied Use of MDL Approximations
A Note on the Applied Use of MDL Approxiations Daniel J. Navarro Departent of Psychology Ohio State University Abstract An applied proble is discussed in which two nested psychological odels of retention
More informationCosine similarity and the Borda rule
Cosine siilarity and the Borda rule Yoko Kawada Abstract Cosine siilarity is a coonly used siilarity easure in coputer science. We propose a voting rule based on cosine siilarity, naely, the cosine siilarity
More informationDetection and Estimation Theory
ESE 54 Detection and Estiation Theory Joseph A. O Sullivan Sauel C. Sachs Professor Electronic Systes and Signals Research Laboratory Electrical and Systes Engineering Washington University 11 Urbauer
More informationExtension of CSRSM for the Parametric Study of the Face Stability of Pressurized Tunnels
Extension of CSRSM for the Paraetric Study of the Face Stability of Pressurized Tunnels Guilhe Mollon 1, Daniel Dias 2, and Abdul-Haid Soubra 3, M.ASCE 1 LGCIE, INSA Lyon, Université de Lyon, Doaine scientifique
More informationarxiv: v1 [stat.ot] 7 Jul 2010
Hotelling s test for highly correlated data P. Bubeliny e-ail: bubeliny@karlin.ff.cuni.cz Charles University, Faculty of Matheatics and Physics, KPMS, Sokolovska 83, Prague, Czech Republic, 8675. arxiv:007.094v
More informationThis article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and
This article appeared in a ournal published by Elsevier. The attached copy is furnished to the author for internal non-coercial research and education use, including for instruction at the authors institution
More informationCourse Notes for EE227C (Spring 2018): Convex Optimization and Approximation
Course Notes for EE227C (Spring 2018): Convex Optiization and Approxiation Instructor: Moritz Hardt Eail: hardt+ee227c@berkeley.edu Graduate Instructor: Max Sichowitz Eail: sichow+ee227c@berkeley.edu October
More informationSampling How Big a Sample?
C. G. G. Aitken, 1 Ph.D. Sapling How Big a Saple? REFERENCE: Aitken CGG. Sapling how big a saple? J Forensic Sci 1999;44(4):750 760. ABSTRACT: It is thought that, in a consignent of discrete units, a certain
More informationA NEW ROBUST AND EFFICIENT ESTIMATOR FOR ILL-CONDITIONED LINEAR INVERSE PROBLEMS WITH OUTLIERS
A NEW ROBUST AND EFFICIENT ESTIMATOR FOR ILL-CONDITIONED LINEAR INVERSE PROBLEMS WITH OUTLIERS Marta Martinez-Caara 1, Michael Mua 2, Abdelhak M. Zoubir 2, Martin Vetterli 1 1 School of Coputer and Counication
More informationIntroduction to Machine Learning. Recitation 11
Introduction to Machine Learning Lecturer: Regev Schweiger Recitation Fall Seester Scribe: Regev Schweiger. Kernel Ridge Regression We now take on the task of kernel-izing ridge regression. Let x,...,
More informationStatistics and Probability Letters
Statistics and Probability Letters 79 2009 223 233 Contents lists available at ScienceDirect Statistics and Probability Letters journal hoepage: www.elsevier.co/locate/stapro A CLT for a one-diensional
More information1 Bounding the Margin
COS 511: Theoretical Machine Learning Lecturer: Rob Schapire Lecture #12 Scribe: Jian Min Si March 14, 2013 1 Bounding the Margin We are continuing the proof of a bound on the generalization error of AdaBoost
More informationWhen Short Runs Beat Long Runs
When Short Runs Beat Long Runs Sean Luke George Mason University http://www.cs.gu.edu/ sean/ Abstract What will yield the best results: doing one run n generations long or doing runs n/ generations long
More informationCurious Bounds for Floor Function Sums
1 47 6 11 Journal of Integer Sequences, Vol. 1 (018), Article 18.1.8 Curious Bounds for Floor Function Sus Thotsaporn Thanatipanonda and Elaine Wong 1 Science Division Mahidol University International
More informationGEE ESTIMATORS IN MIXTURE MODEL WITH VARYING CONCENTRATIONS
ACTA UIVERSITATIS LODZIESIS FOLIA OECOOMICA 3(3142015 http://dx.doi.org/10.18778/0208-6018.314.03 Olesii Doronin *, Rostislav Maiboroda ** GEE ESTIMATORS I MIXTURE MODEL WITH VARYIG COCETRATIOS Abstract.
More informationSEISMIC FRAGILITY ANALYSIS
9 th ASCE Specialty Conference on Probabilistic Mechanics and Structural Reliability PMC24 SEISMIC FRAGILITY ANALYSIS C. Kafali, Student M. ASCE Cornell University, Ithaca, NY 483 ck22@cornell.edu M. Grigoriu,
More informationarxiv: v1 [cs.ds] 3 Feb 2014
arxiv:40.043v [cs.ds] 3 Feb 04 A Bound on the Expected Optiality of Rando Feasible Solutions to Cobinatorial Optiization Probles Evan A. Sultani The Johns Hopins University APL evan@sultani.co http://www.sultani.co/
More informationA general forulation of the cross-nested logit odel Michel Bierlaire, Dpt of Matheatics, EPFL, Lausanne Phone: Fax:
A general forulation of the cross-nested logit odel Michel Bierlaire, EPFL Conference paper STRC 2001 Session: Choices A general forulation of the cross-nested logit odel Michel Bierlaire, Dpt of Matheatics,
More informationA proposal for a First-Citation-Speed-Index Link Peer-reviewed author version
A proposal for a First-Citation-Speed-Index Link Peer-reviewed author version Made available by Hasselt University Library in Docuent Server@UHasselt Reference (Published version): EGGHE, Leo; Bornann,
More informationCOS 424: Interacting with Data. Written Exercises
COS 424: Interacting with Data Hoework #4 Spring 2007 Regression Due: Wednesday, April 18 Written Exercises See the course website for iportant inforation about collaboration and late policies, as well
More informationOnline Publication Date: 19 April 2012 Publisher: Asian Economic and Social Society
Online Publication Date: April Publisher: Asian Econoic and Social Society Using Entropy Working Correlation Matrix in Generalized Estiating Equation for Stock Price Change Model Serpilılıç (Faculty of
More informationA Poisson process reparameterisation for Bayesian inference for extremes
A Poisson process reparaeterisation for Bayesian inference for extrees Paul Sharkey and Jonathan A. Tawn Eail: p.sharkey1@lancs.ac.uk, j.tawn@lancs.ac.uk STOR-i Centre for Doctoral Training, Departent
More informationANALYTICAL INVESTIGATION AND PARAMETRIC STUDY OF LATERAL IMPACT BEHAVIOR OF PRESSURIZED PIPELINES AND INFLUENCE OF INTERNAL PRESSURE
DRAFT Proceedings of the ASME 014 International Mechanical Engineering Congress & Exposition IMECE014 Noveber 14-0, 014, Montreal, Quebec, Canada IMECE014-36371 ANALYTICAL INVESTIGATION AND PARAMETRIC
More informationA Bernstein-Markov Theorem for Normed Spaces
A Bernstein-Markov Theore for Nored Spaces Lawrence A. Harris Departent of Matheatics, University of Kentucky Lexington, Kentucky 40506-0027 Abstract Let X and Y be real nored linear spaces and let φ :
More informationChaotic Coupled Map Lattices
Chaotic Coupled Map Lattices Author: Dustin Keys Advisors: Dr. Robert Indik, Dr. Kevin Lin 1 Introduction When a syste of chaotic aps is coupled in a way that allows the to share inforation about each
More informationComputational and Statistical Learning Theory
Coputational and Statistical Learning Theory Proble sets 5 and 6 Due: Noveber th Please send your solutions to learning-subissions@ttic.edu Notations/Definitions Recall the definition of saple based Radeacher
More informationThe Hilbert Schmidt version of the commutator theorem for zero trace matrices
The Hilbert Schidt version of the coutator theore for zero trace atrices Oer Angel Gideon Schechtan March 205 Abstract Let A be a coplex atrix with zero trace. Then there are atrices B and C such that
More informationNumerical Studies of a Nonlinear Heat Equation with Square Root Reaction Term
Nuerical Studies of a Nonlinear Heat Equation with Square Root Reaction Ter Ron Bucire, 1 Karl McMurtry, 1 Ronald E. Micens 2 1 Matheatics Departent, Occidental College, Los Angeles, California 90041 2
More informationVariance Reduction. in Statistics, we deal with estimators or statistics all the time
Variance Reduction in Statistics, we deal with estiators or statistics all the tie perforance of an estiator is easured by MSE for biased estiators and by variance for unbiased ones hence it s useful to
More informationSharp Time Data Tradeoffs for Linear Inverse Problems
Sharp Tie Data Tradeoffs for Linear Inverse Probles Saet Oyak Benjain Recht Mahdi Soltanolkotabi January 016 Abstract In this paper we characterize sharp tie-data tradeoffs for optiization probles used
More informationHermite s Rule Surpasses Simpson s: in Mathematics Curricula Simpson s Rule. Should be Replaced by Hermite s
International Matheatical Foru, 4, 9, no. 34, 663-686 Herite s Rule Surpasses Sipson s: in Matheatics Curricula Sipson s Rule Should be Replaced by Herite s Vito Lapret University of Lublana Faculty of
More informationAlireza Kamel Mirmostafaee
Bull. Korean Math. Soc. 47 (2010), No. 4, pp. 777 785 DOI 10.4134/BKMS.2010.47.4.777 STABILITY OF THE MONOMIAL FUNCTIONAL EQUATION IN QUASI NORMED SPACES Alireza Kael Mirostafaee Abstract. Let X be a linear
More informationTesting Properties of Collections of Distributions
Testing Properties of Collections of Distributions Reut Levi Dana Ron Ronitt Rubinfeld April 9, 0 Abstract We propose a fraework for studying property testing of collections of distributions, where the
More informationIN modern society that various systems have become more
Developent of Reliability Function in -Coponent Standby Redundant Syste with Priority Based on Maxiu Entropy Principle Ryosuke Hirata, Ikuo Arizono, Ryosuke Toohiro, Satoshi Oigawa, and Yasuhiko Takeoto
More informationExact Linear Likelihood Inference for Laplace
Exact Linear Likelihood Inference for Laplace Prof. N. Balakrishnan McMaster University, Hamilton, Canada bala@mcmaster.ca p. 1/52 Pierre-Simon Laplace 1749 1827 p. 2/52 Laplace s Biography Born: On March
More informationResearch in Area of Longevity of Sylphon Scraies
IOP Conference Series: Earth and Environental Science PAPER OPEN ACCESS Research in Area of Longevity of Sylphon Scraies To cite this article: Natalia Y Golovina and Svetlana Y Krivosheeva 2018 IOP Conf.
More informationUsing EM To Estimate A Probablity Density With A Mixture Of Gaussians
Using EM To Estiate A Probablity Density With A Mixture Of Gaussians Aaron A. D Souza adsouza@usc.edu Introduction The proble we are trying to address in this note is siple. Given a set of data points
More informationInference in the Presence of Likelihood Monotonicity for Polytomous and Logistic Regression
Advances in Pure Matheatics, 206, 6, 33-34 Published Online April 206 in SciRes. http://www.scirp.org/journal/ap http://dx.doi.org/0.4236/ap.206.65024 Inference in the Presence of Likelihood Monotonicity
More informationA remark on a success rate model for DPA and CPA
A reark on a success rate odel for DPA and CPA A. Wieers, BSI Version 0.5 andreas.wieers@bsi.bund.de Septeber 5, 2018 Abstract The success rate is the ost coon evaluation etric for easuring the perforance
More informationQuantum algorithms (CO 781, Winter 2008) Prof. Andrew Childs, University of Waterloo LECTURE 15: Unstructured search and spatial search
Quantu algoriths (CO 781, Winter 2008) Prof Andrew Childs, University of Waterloo LECTURE 15: Unstructured search and spatial search ow we begin to discuss applications of quantu walks to search algoriths
More informationInspection; structural health monitoring; reliability; Bayesian analysis; updating; decision analysis; value of information
Cite as: Straub D. (2014). Value of inforation analysis with structural reliability ethods. Structural Safety, 49: 75-86. Value of Inforation Analysis with Structural Reliability Methods Daniel Straub
More informationAn Introduction to Meta-Analysis
An Introduction to Meta-Analysis Douglas G. Bonett University of California, Santa Cruz How to cite this work: Bonett, D.G. (2016) An Introduction to Meta-analysis. Retrieved fro http://people.ucsc.edu/~dgbonett/eta.htl
More informationLower Bounds for Quantized Matrix Completion
Lower Bounds for Quantized Matrix Copletion Mary Wootters and Yaniv Plan Departent of Matheatics University of Michigan Ann Arbor, MI Eail: wootters, yplan}@uich.edu Mark A. Davenport School of Elec. &
More information