STATISTICS RESEARCH ARTICLE. s 4
|
|
- Suzanna Theodora Wright
- 6 years ago
- Views:
Transcription
1 Chatzopoulos & Kolyva-Machrea, Cogent Mathematcs & Statstcs (018), 5: STATISTICS RESEARCH ARTICLE D-optmal saturated factoral desgns St. A. Chatzopoulos 1 * and F. Kolyva-Machrea 1 Receved: 0 October 017 Accepted: March 018 Frst Publshed: 9 March 018 *Correspondng author: St. A. Chatzopoulos, Department of Mathematcs, Arstotle Unversty, Thessalonk 51, Greece E-mal: cstavros@math.auth.gr Revewng edtor: Guohua Zou, Chnese Academy of Scences, Chna Addtonal nformaton s avalable at the end of the artcle Abstract: In ths paper, Resoluton III saturated s 1 s, s s 3 s factoral desgns and specally the cases (s k) s, s k, k = 0, 1 are studed, n order to obtan D-optmal plans. Subjects: Scence; Mathematcs & Statstcs; Statstcs & Probablty; Statstcs; Mathematcal Statstcs; Statstcal Computng; Statstcal Theory & Methods Keywords: D-optmal desgns; saturated desgns; desgn matrx AMS subject classfcatons: Prmary 6K05; 6K15 ; Secondary 11C0; 6J05 1. Introducton Saturated factoral plans s a very nterestng ssue n theory of exeprmental desgns, snce the reduced number of observatons s very usefull n practse especally n screenng experments, where are used to determne whch of many factors affects the measure of pertnent qualty characterstcs. In saturated desgns the number of observaton s equal to the number of parameters, so all degrees of freedom are consumed by the estmaton of parameters, leavng no degrees of freedom for error varance estmaton. The purpose of ths paper s to gve saturated resoluton III desgns, mnmzng the generalzed varance of the man effects and the general mean, that s, D-optmal desgns. In recent years, there has been a consderable nterest n optmal saturated man effect desgns wth two or three factors. Mukerjee et al. (1986) and Kraft (1990) showed all two-factor desgns are equvalent wth respect to D-optmalty crteron. Later Mukerjee and Snha (1990) ABOUT THE AUTHORS The problem of fndng optmal desgns under dfferent types of crtera preoccupes many researchers the last decades. Most of the work on constructng optmal desgns for the estmaton of parameters n fractonal factorals s concentrated on factors at two levels. Chatzopoulos, Kolyva-Machera, and Chatterjee (009), studed the optmalty of desgns whch are obtaned by addng p runs to an orthogonal array for experments nvolvng m factors each at s levels. Chatterjee, Kolyva-Machera, and Chatzopoulos (011), consdered the ssue of optmalty of fractonal factoral experments nvolvng m factors each at two levels. Percleous, Chatzopoulos, Kolyva-Machera and Kounas, study the problem of estmatng the standardzed lnear and quadratc contrasts n fractonal factorals wth k factors, each at 3 levels, when the number of runs or assembles s N = 3 and ntroduced a dfferent noton of Balanced Arrays. Chatzopoulos and Kolyva-Machera (005), studed the saturated m 1 m m 3 desgns and Chatzopoulos & Kolyva- Machera (008), consdered the problem of fndng D-optmal saturated m m 3 desgns. PUBLIC INTEREST STATEMENT An ssue of nterestng n expermental desgns s the saturated desgns. An expermental desgn s called saturated f all the degrees of freedom are consumed by the estmaton of the parameters wthout leavng degrees of freedom for error varance estmaton. The saturated factoral desgns, where the nterest s to estmate the general mean and the man effects whle all hgher order nteractons are neglgble (resoluton III plans), are commonly used n screenng experments. In recent years, there has been a consderable nterest n optmal saturated man effect desgns. Most researchers have dealt wth the case where two or three factors are nvolved n the experment on two levels. The problem s dfferent and becomes more dffcult when three or four factors are nvolved n the experment on three or more levels. 018 The Author(s). Ths open access artcle s dstrbuted under a Creatve Commons Attrbuton (CC-BY).0 lcense. Page 1 of 13
2 Chatzopoulos & Kolyva-Machrea, Cogent Mathematcs & Statstcs (018), 5: consdered, for the two-factor case, the optmalty results on almost saturated man effect desgns. Pesotan and Raktoe (1988) worked also n the specal case for s factorals and a subclass of s 3 factorals. Chatterjee and Mukerjee (1993) were the frst who attempted to extend the two factor results to three factors. They consder s, s, s 3 s factoral to derve D-optmal saturated man effect desgns. Later Chatterjee and Narasmhan (00), usng technques from Graph Theory and Combnatorcs, clamed about the upper bound of the determnant of the saturated 3 s, s 3, s 3 s factorals when s s odd. Chatzopoulos and Kolyva-Machera (006) extend the results concernng D-optmal saturated man effect desgns for s to 3 s factorals, when 3 s 6 and s 3 s. Karaganns and Moyssads (005) and Karaganns and Moyssads (008) extend the Graph theoretc approach of Chatterjee and Narasmhan (00) and the results of Chatzopoulos and Kolyva-Machera (006), and gve the D-optmal saturated 3 s, s 3, s 3 s desgns. In ths paper, we study the D-optmalty for saturated s 1 s factorals. Moreover, we gve the upper bound of the determnant for the (s k) s, s k, k = 0, 1, saturated desgns and the correspondng desgn, whch attans ths bound. The paper s organzed as follows. Some notatons and prelmnares are frst presented n Secton. Secton 3 deals wth the man results of ths paper.. Notatons and prelmnares In ths paper, we follow the same notatons as n Chatzopoulos and Kolyva-Machera (006) adapted for four factors. Let us consder the setup of an s 1 s,s s 3 s saturated factoral experment, nvolvng four factors F 1, F, F 3 and F appearng at s 1, s, s 3 and s levels, respectvely, wth N = s 1 + s + s 3 runs. For 1 let the levels of F be denoted by τ and coded as 0, 1, s 1. Our nterest s to fnd D-optmal resoluton III desgns. There are altogether s 1 s s 3 s treatment combnatons denoted by τ 1 τ τ 3 τ, that wll hereafter be assumed to be lexcographcally ordered. Let, for 1, 1 be the s 1 vector wth each element unty, I the dentty matrx of order s, denotes the Kronecker product of matrces and P be an (s 1) s matrx such that (s ( 1 ) 1, P ) s orthogonal (A denotes the transpose of matrx A). The usual fxed effect model under the absence of nteractons s Y = Wβ + ε, where Y s the response vector of the experment, ε s the vector of uncorrelated random errors wth zero mean and the same varance σ and β s the vector of unknown parameters, s consder. In our case β =(μ, β, 1 β, β, 3 β ), where μ s the unknown general mean and the elements of the (s 1) 1 vectors β are unknown parameters representng a full set of mutually orthogonal contrasts belongng to the man effects F and W =[ , W 1, W, W 3, W ], where W 1 = P , W = 1 1 P 1 1 3, W = P 1 3 and W = P. It s easy to see that the D-optmal desgn does not depend on the choce of P, 1. Followng Mukerjee and Snha (1990) let X 0 =[ , X 1, X, X 3, X ], where X 1 = I , X = 1 1 I 1 3 1, X 3 = I 3 1 and X = I. We denote X (1), = 1,, 3 the matrces obtaned by deletng the frst column of X, = 1,, 3. Consder the u (s 1 + s + s 3) matrx U, whch s a submatrx of X 0 gven by U =[X (1), 1 X(1), X(1), X 3 ], whch has full column rank. The u rows of matrx U lke those of W, correspond to the lexcographcally ordered treatment combnatons. Moreover the columns of U span those of X 0 and hence those of W, whch also has full column rank. Hence, one may obtan W = UH, where matrx H s a nonsngular matrx of order s 1 + s + s 3. For any desgn d n the class of the saturated resoluton III desgns wth N = s 1 + s + s 3 runs, the desgn matrx s W d = H, where s a square matrx of order s 1 + s + s 3 such that for 1 j s 1 + s + s 3 f the -th run n d s gven by the treatment combnaton τ 1 τ τ 3 τ then the j-th row of s the row of U correspondng to the treatment combnaton τ 1 τ τ 3 τ. A desgn d s sad to be D-optmal n the class, f t maxmzes the Page of 13
3 Chatzopoulos & Kolyva-Machrea, Cogent Mathematcs & Statstcs (018), 5: quantty det(w W d d ). Snce matrx H s nonsngular a desgn s D-optmal f t maxmzes the quantty det( ), where: =[Z (1) 1, Z(1), Z(1) 3, Z ]. (1) The matrces Z (1), 1 3 and Z are obtan.ed from the matrces X (1) and X n a smlar way, as s obtaned from U. Defnton.1 For 1 3, f the -th factor enters the experment at level 0 then the correspondng row of the matrx Z (1) s a row vector wth s 1 elements zero. On the other hand f the -th factor enters the experment at level p, 1 p (s 1), then the correspondng row of the matrx Z (1) equals the p-th row of the dentty matrx of order (s 1). Smlarly, f the fourth factor enters the experment at level p, 0 p s 1, then the correspondng row of the matrx Z equals to the (p + 1)-th row of the dentty matrx I s. Let n p, 0 p (s 1), denote the number of these rows. It holds that (s 1) N = n p, = 1,, 3,. Defnton. For 1 j and 0 p (s 1), 0 q (s j 1), let n pq j, denote the number of runs where the -th factor appears at level p and the j-th factor appears at level q. It holds that n p = p=0 (s j 1) q=0 n pq j, for j = 1,, 3,, j, (s 1) n q = n pq, for = 1,, 3,, j, j j N = p=0 (s 1) p=0 (s j 1) q=0 n pq j, for 1 j. Defnton.3 For 1 j k and 0 p (s 1), 0 q (s j 1), 0 r (s k 1) let n pqr, denote the number of runs where the -th factor appears at p level, the j-th factor appears at level q and jk the k-th factor appears at level r. It holds that (s k 1) n pq = n pqr, for k = 1,, 3,, k j, j jk n pr k n qr jk N = r=0 (sj 1) = q=0 (s 1) = p=0 (s 1) p=0 n pqr, for j = 1,, 3,, k j k, jk n pqr, for = 1,, 3,, j k j, jk (s j 1) q=0 (s k 1) r=0 n pqr, for 1 j k. jk Remark.1 It holds that n p desgn has full column rank. 1, 1, 0 p (s 1), snce the desgn matrx of a saturated Remark. By the choce of the labels for the levels one can always assume, wthout loss of generalty (w.l.g), that n 0 = max{n p,0 p s 1, 1 }. The followng lemmas are crucal for the man results of our paper and can be founded n Chatterjee and Mukerjee (1993) and Chatzopoulos and Kolyva-Machera (006). Page 3 of 13
4 Chatzopoulos & Kolyva-Machrea, Cogent Mathematcs & Statstcs (018), 5: Lemma.1 Consder the saturated s 1 s desgn e, s, wth N = s 1 + s 1 runs and correspondng matrx X e. It holds that det(x e ) = 1. () Proof See Chatterjee and Mukerjee (1993). Lemma. Consder the saturated s desgns d (), s 3 s, wth N = s runs and correspondng matrx U (). It holds that d det(u () d ) s. (3) Proof See Chatterjee and Mukerjee (1993). Lemma.3 Consder the saturated s 1 s s k desgn d. If s and n p = 1 for some 0 p s 1 then det( ) = det( ), where d s a saturated s 1 (s 1) s k desgn. Proof See Chatzopoulos and Kolyva-Machera (006), lemma.1. Corollary.1 Consder the saturated s 1 s, s s 3 s desgn d wth N = s 1 + s + s 3 runs. If w = s (s 1 + s 3) 0, then usng the pgeonhole prncple we can easly verfy that n p = 1, for some 0 p s 1 at least w tmes. Applyng, w tmes, lemma.3, we get det( ) = det( ), where d s a saturated s 1 s (s 1 + s 3) desgn. Remark.3 For s 1 = s = we have to study only the cases where 0 s s 3 1, that s the cases s, s and (s 1) s, s 3. Lemma. Let d be a saturated s 1 s, s 3 s desgn. If det( ) = s 1 p=1 np and n 0 n 1 n (s 1) n 0 1, = 1,, 3, then d s D-optmal. Proof See Chatzopoulos and Kolyva-Machera (006), theorem Man results Lemma 3.1 The determnant of the matrx =[Z (1), 1 Z(1), Z(1), Z 3 ], gven n (1), whch corresponds to a s 1 s, s s 3 s saturated factoral desgn d s left nvarant by nterchangng the levels of the factors. Proof For the fourth factor, we can nterchange the columns whch correspond to two levels and the proof s obvous. Smlarly, for the nonzero levels of the frst, second and the thrd factor, nterchangng the columns p and q, the levels (p 1) and (q 1) are nterchanged. Moreover, for the frst (or second or thrd) factor, addng all the columns of matrx Z 1 (or 1 Z1 or Z1 3 ) to the column whch corresponds to level p, subtractng the sum of all columns of matrx Z and multplyng the resultng column by ( 1) the levels 0 and p are nterchanged. Lemma 3. Let d be a s 1 s, s s 3 s saturated factoral desgn wth correspondng matrx =[Z (1), 1 Z(1), Z(1), Z 3 ] as gven n (1). Let npqr = w > 1 for some 1, j, k and some jk 0 p s 1, 0 q s j 1, 0 r s k 1. Then det(u d (13)) for(, j, k) =(1,, 3), det(u det( ) = d (1)) for(, j, k) =(1,, ), det( (13)) for(, j, k) =(1, 3, ), det( (3)) for(, j, k) =(, 3, ), Page of 13
5 Chatzopoulos & Kolyva-Machrea, Cogent Mathematcs & Statstcs (018), 5: where d (13), d (1), d (13) and d (3) s s 1 s (s w + 1), s 1 s (s 3 w + 1) s, s 1 (s w + 1) s 3, (s 1 w + 1) s, saturated factoral desgn, respectvely. Proof Let as assume that n pqr = w > 1, 0 p s 1, 0 q s 1, 0 r s , whch means that the saturated desgn s 1 s contans the runs pqrx 1, pqrx,, pqrx w. By subtractng the row correspondng to run pqrx 1 from the other rows whch correspond to runs pqrx,, pqrx w, addng the columns correspondng to levels x,, x w to the column correspondng to level x 1 and expandng det( ) along the (w 1) rows whch contan levels x,, x w, we get det( ) = det( (13)), where d (13) s s 1 s (s w + 1) saturated desgn. The proof s smlar for n pqr = w, npqr = w 1 13 and n pqr = w D-optmalty of (s 1) s saturated desgns Lemma 3.3 Consder the saturated (s 1) s desgn d wth N = s runs and correspondng matrx as gven n (1). For the D-optmal desgn t holds that: n p = s, = 1,, p = 0, 1. n 0 = 3 n1 = 3, 3 np =, p (s ). 3 =, 0 p (s 1). n p Proof Expandng det( ) along ts frst column, we have that det( ) = n1 1 det(u =1 d ), where d (), () = 1,,, n 1 are (s 1) s saturated desgns wth correspondng matrces U 1 d. Let det(u () d ) = max { det( ()), = 1,,, n 1 }. Then det( ) n 1 det( ). () From lemma 3.1, by nterchangng the levels 0 and 1 of the frst factor, t also holds det( ) n 0 det( ). (5) From ()-(5), we get det( ) n0 1 +n1 1 det(u d ) = s det( ). Accordng to lemma., n order to fnd the D-optmal desgn d, t must hold that n 0 n 1 n s 1 n 0 1, =, 3,, whch mples n 0 = n1 = s, n0 = 3 n1 = 3, 3 np =, p (s ), 3 np =, 0 p (s 1). Expandng det(u d ) along ts second column, and followng the same procedure we get n 0 = 1 n1 = s. 1 Lemma 3. Consder the saturated (s 1) s desgn d wth N = s runs and correspondng matrx as gven n (1). For the D-optmal desgn t holds that: n 0q = n1q = 1, = 1,, 0 q (s 1). Proof get n 0q 1 Suppose, w.l.g, that n 0p = for some 0 p (s 1). Then from the pgeonhole prncple, we 1 = for some q, 0 p q (s 1). From lemmas 3.1 and 3., we may assume that n a desgn d the followng treatment combnatons exst: 000τ p, 01τj 3 τp, 11τk 3 τq, 10τ 3 τq. Consder now matrx U, d whch corresponds to the desgn d, as gven n (1). Subtract the row correspondng to the treatment combnato0τ p from the row correspondng to the treatment combnaton 01τj 3 τp. The row correspondng to the treatment combnaton 01τ j 3 τp s now r =(0, 1, 0,, 0, 1, 0,, 0, 0,,0), where the second ace s at the (j + )-th column of r, that s at the j-th column of Z (1). Then, subtract the row r 3 from the row correspondng to the treatment combnaton 11τ k 3 τq. Contnue by addng the column of Z (1), whch corresponds to 3 τj to the column of Z(1) whch corresponds to 3 3 τk 3. Consequently, treatment combnaton 10τ k 3 τp s now n the poston of treatment combnaton 11τk 3 τp. Add the row correspondng to the treatment combnato0τ p to row r. The resultng row corresponds to the treatment combnaton 01τ j 3 τp. Hence, the desgn d contans the treatment combnatons 000τp, 01τj 3 τp, 10τk 3 τq, 10τ 3 τq, whch mples n10q =. Then, proceedng as n lemma 3., we have that det(u ) = det(u 1 d d ), Page 5 of 13
6 Chatzopoulos & Kolyva-Machrea, Cogent Mathematcs & Statstcs (018), 5: where d s (s ) (s 1) saturated desgn. Corollary 3.1 For the D-optmal saturated (s 1) s desgn d wth N = s runs, f there exsts the treatment combnatoτ 3 τ (01τ 3 τ ) then there exsts the treatment combnaton 11τ 3 τ (10τ 3 τ ). Lemma 3.5 Consder the saturated (s 1) s desgn d wth N = s runs and correspondng matrx as gven n (1). For the D-optmal desgn t holds that: n pq = 1, = 1,, 0 p 1, q (s ). 3 Proof The proof s smlar as n lemma 3.. Corollary 3. Let us now consder the saturated (s 1) s desgn d wth N = s runs. Then, from lemma 3. and usng the pgeonhole prncple, we get: n pq 1, 0 p, q (s 1), 3 n pq =, 1, 0 p 1, 0 q 1. 3 Corollary 3.3 For the D-optmal saturated (s 1) s desgn d wth N = s runs t holds that: n pq 1 = s f s 0 mod, 0 p, q 1, = 1 n11 = s+1 1 = 1 n11 = s 1 1 and n 01 = 1 n10 = s 1 1 and n 01 = 1 n10 = s+1 1 or } f s 1 mod. Lemma 3.6 Consder the saturated (s 1) s desgn d wth N = s runs and correspondng matrx as gven n (1). For the D-optmal desgn t holds that: 0 13 = n = n = n = n = n = 1. (6) Proof From lemma 3.3, we have n p = 3, p = 0, 1 and from lemma 3. and corollary 3. the D-optmal 3 desgn ncludes one of the followng sets of treatment combnatons: (00pτ, 01pτj, 11pτk) or (00pτ, 01pτ j, 10pτk) or (00pτ, 10pτj, 11pτk) or (11pτ, 01pτj, 10pτk ). Applyng lemma 3.1, we can always choose, w.l.g. the treatment combnato0τ, 010τj, 110τk. Ths choce, usng pgeonhole prncple, mples the exstence of the treatment combnatons 001t q, 111tr, 101tt and the proof of (6) s obvous. Theorem 3.1 Consder the saturated (s 1) s desgn d wth N = s runs and correspondng matrx as gven n (1). It holds that: det( ) { s s 1 fs 0 mod, fs 1 mod. (7) Proof Let u = = 1 n11 1. So, from corollary 3.3, we have u = s f s 0 mod or u =(s + 1) f s 1 mod. Moreover, from lemmas and corollares , w.l.g., the D-optmal saturated desgn (s 1) s can be wrtten as: Page 6 of 13
7 Chatzopoulos & Kolyva-Machrea, Cogent Mathematcs & Statstcs (018), 5: Let 0 1 k be a 1 k vector wth all elements equal to zero, I k be the dentfy matrx of order k. For m < k, t can be easly seen that: I k = e 1k e k e kk = e 1m e mm 0 1 m 0 1 m 0 1 (k m) 0 1 (k m) e 1(k m) e (k m)(k m), (8) Matrx as gven n (1), can be wrtten as: Page 7 of 13
8 Chatzopoulos & Kolyva-Machrea, Cogent Mathematcs & Statstcs (018), 5: Usng relaton (8), and after permutaton of columns matrx can be wrtten as: Now subtract the (u)-th column from the (u + 1)-th column and add the last s u columns to the (u + 1)-th column. Matrx, as gven n (9), can be wrtten as: = ( Ud1 0 u (s u) A U ). Matrx 1 s the desgn matrx of the saturated u u desgn d 1 wth N 1 = u runs. Matrx U s not desgn matrx as ts frst column s (,,, 0,,0), but det(u ) = det( ), where matrx s the desgn matrx of the saturated (s u) (s u) desgn d wth N = (s u) runs. Hence, det( ) = det(1 ) det( ). From (3), we get det( ) u (s u). Recallng that u = s f s 0 mod, or u =(s + 1) f s 1 mod, we get that (7) holds. Theorem 3. Let u =(s 3 + 1) f s 1 mod, or u = s f s 0 mod. The saturated, s > s 3 desgn d wth the followng N = s 3 + s + 1 treatment combnatons 00 (0 u 1), 110(u 1), 11( + 1) (0 u ), 101u, 10( + 1) (u s 3 1), 010s 3, 01 (u s 3 1), 01(s 3 1) (s s 1) s a D-optmal desgn n the class of all, s > s 3 saturated desgns. Proof The proof s obvous from lemma.3, lemmas and theorem D-optmalty of s saturated desgns Lemma 3.7 Consder the saturated s desgn d wth N = s + 1 runs and correspondng matrx as gven n (1). For the D-optmal desgn, t holds that: n 0 = s + 1 and n 1 = s, or n 0 = s and n 1 = s + 1, = 1,. n 0 = 3, n p =, = 3,, 1 p s 1. Proof The proof s smlar as lemma 3.3. Page 8 of 13
9 Chatzopoulos & Kolyva-Machrea, Cogent Mathematcs & Statstcs (018), 5: Corollary 3. For the saturated s desgn d wth N = s + 1 runs, from lemma 3. and usng the pgeonhole prncple we get: n pq s + 1, 0 p, q 1, 1 n pq, 0 p, q s 1, 3 n pq, 1, 0 p 1, 3 j, 0 q s 1. j Lemma 3.8 Consder the saturated s desgn d wth N = s + 1 runs and correspondng matrx as gven n (1). For the D-optmal desgn, t holds that: n 0q = n1q = 1, = 1,, 1 q (s 1). j =, or n 10 j =, = 1,, 3 j. (10) (11) Proof For the proof of relaton (10) see lemma 3.. The proof of relaton (11) s obvous, snce n 1 = s + 1 or n 0 = s + 1. Corollary 3.5 For the D-optmal saturated s desgn d wth N = s + 1 runs t holds that: 1 = n01 1 = n10 1 = s 1 = n01 1 = n11 1 = s 1 = n10 1 = n11 1 = s n 01 1 = n10 1 = n11 1 = s and n 11 = s + 1 or 1 and n 10 = s + 1 or 1 and n 01 = s + 1 or f s 0 mod. 1 and = s = n01 1 = n10 1 = s+1 1 = n01 1 = n11 1 = s+1 1 = n10 1 = n11 1 = s+1 n 01 1 = n10 1 = n11 1 = s+1 and n 11 1 = s 1 and n 10 1 = s 1 and n 01 1 = s 1 and 1 = s 1 or or f s 1 mod. or Theorem 3.3 Consder the saturated s desgn d wth N = s + 1 runs and correspondng matrx as gven n (1). It holds that: det( ) s(s + 1). Proof If s 0 mod, then u = s = = 1 n01 = 1 n11 1 and n10 1 = u + 1, whle f s 1 mod, then u =(s 1) = n 01 1 and n00 = 1 n10 = 1 n11 1 = u + 1. From lemmas and corollares , the D- optmal saturated s desgn, w.l.g., can be wrtten as: (1) Page 9 of 13
10 Chatzopoulos & Kolyva-Machrea, Cogent Mathematcs & Statstcs (018), 5: By nterchangng the levels 0 and (s 1) of the -th factor, accordng to lemma 3.1, the determnant of the matrx s left nvarant. Hence the D-optmal saturated s desgn, w.l.g., can be wrtten as: Matrx as gven n (1), can be wrtten as: Page 10 of 13
11 Chatzopoulos & Kolyva-Machrea, Cogent Mathematcs & Statstcs (018), 5: Usng relaton (8), and after a sutable permutaton of columns of the matrx, n order to make the left bottom block of the matrx a zero matrx, matrx can be wrtten as: Analytcally, the permutaton s: the frst u columns of matrx are the frst u columns of matrx Z (1) 3, columns (u + 1) u of matrx are the frst u columns of matrx Z, (u + 1)-th column of matrx s matrx Z (1) 1, (u + )-th column of matrx columns of matrx are the rest columns of matrces Z (1) 3 and Z. s matrx Z(1), whle the remanng (s + 1) (u + ) Page 11 of 13
12 Chatzopoulos & Kolyva-Machrea, Cogent Mathematcs & Statstcs (018), 5: Now subtract the (u + )-th column from the (u + 1)-th column. Then, matrx s a block trangular matrx. It holds that det( ) = det(u 1 ) det( ), where matrx U 1 s a (u + 1) (u + 1) matrx, whch does not correspond to any desgn and matrx s the desgn matrx of the saturated (s u) (s u) desgn d wth N = (s u) runs. From (3), we get det (s u). Moreover, expandng det(u 1 ) along ts last column we get that det(u 1 ) u+1 j=1 det(x ej ), where, after some manpulatons, X ej correspond to two factor saturated desgns e j, wth, accordng to lemma.1, det(x ej ) = 1. Hence, det( ) (u + 1)(s u). Recallng that u = s f s 0 mod, or u =(s 1) f s 1 mod, we get that (11) holds. Theorem 3. Let u = s f s 0 mod, or u =(s 1) f s 1 mod. The saturated s, s desgn d wth the followng N = s + 1 runs 100(u 1), 101(s 1), 10( + 1) (1 u ), 10u0, 01 (1 u 1), 01u(s 1), 110u, 11 (u + 1 s ), 11(s 1)0, 0000, 00( + 1) (u s ), s a D-optmal desgn n the class of all s, s saturated desgns. Proof From lemma.3, lemmas 3.1, 3., 3.7, 3.8 and theorem 3.3, the proof s obvous. Acknowledgements We would lke to thank the referees and the journal edtoral team for provdng valuable advce that mproved the qualty of the orgnal manuscrpt. Fundng The authors receved no drect fundng for ths research. Author detals St. A. Chatzopoulos 1 E-mal: cstavros@math.auth.gr F. Kolyva-Machrea 1 E-mal: fkolyva@math.auth/gr 1 Department of Mathematcs, Arstotle Unversty, Thessalonk 51, Greece. Ctaton nformaton Cte ths artcle as: D-optmal s3 s saturated factoral desgns, St. A. Chatzopoulos & F. Kolyva-Machrea, Cogent Mathematcs & Statstcs (018), 5: References Chatterjee, K., & Mukerjee, R. (1993). D-optmal saturated man effect plans for s factorals. Journal of Combnatorcs, Informaton & System Scences, 18, Chatterjee, K., & Narasmhan, G. (00). Graph theoretc technques n D-optmal desgn problems. Journal of Statstcal Plannng and Inference, 10, Chatzopoulos, S. A., & Kolyva-Machera, F. (006). Some D-optmal saturated desgns for 3 m m 3 factorals. Journal of Statstcal Plannng and Inference, 136, Karaganns, V., & Moyssads, C. (005). Constructon of D-optmal s 1 s factoral desgns usng graph theory. Metrka, 6, Karaganns, V., & Moyssads, C. (008). A graphcal constructon of the D-optmal saturated, 3 s man effect, factoral desgn. Journal of Statstcal Plannng and Inference, 138, Kraft, O. (1990). Some matrx representatons occurrng n two factor models. In R. R. Bahadur (Ed.), Probablty, statstcs and desgn of experments (pp ). New Delh: Wley Eastern. Mukerjee, R., Chatterjee, K., & Sen, M. (1986). D-optmalty of a class of saturated man effect plans and alled results. Statstcs, 17(3), Mukerjee, R., & Snha, B. K. (1990). Almost saturated D-optmal man effect plans and alled results. Metrka, 37, Pesotan, H., & Raktoe, B. L. (1988). On nvarance and randomzaton n factoral desgns wth applcatons to D-optmal man effect desgns of the symmetrcal factoral. Journal of Statstcal Plannng and Inference, 19, Page 1 of 13
13 Chatzopoulos & Kolyva-Machrea, Cogent Mathematcs & Statstcs (018), 5: The Author(s). Ths open access artcle s dstrbuted under a Creatve Commons Attrbuton (CC-BY).0 lcense. You are free to: Share copy and redstrbute the materal n any medum or format Adapt remx, transform, and buld upon the materal for any purpose, even commercally. The lcensor cannot revoke these freedoms as long as you follow the lcense terms. Under the followng terms: Attrbuton You must gve approprate credt, provde a lnk to the lcense, and ndcate f changes were made. You may do so n any reasonable manner, but not n any way that suggests the lcensor endorses you or your use. No addtonal restrctons You may not apply legal terms or technologcal measures that legally restrct others from dong anythng the lcense permts. Cogent Mathematcs & Statstcs (ISSN: ) s publshed by Cogent OA, part of Taylor & Francs Group. Publshng wth Cogent OA ensures: Immedate, unversal access to your artcle on publcaton Hgh vsblty and dscoverablty va the Cogent OA webste as well as Taylor & Francs Onlne Download and ctaton statstcs for your artcle Rapd onlne publcaton Input from, and dalog wth, expert edtors and edtoral boards Retenton of full copyrght of your artcle Guaranteed legacy preservaton of your artcle Dscounts and wavers for authors n developng regons Submt your manuscrpt to a Cogent OA journal at Page 13 of 13
ISSN: ISO 9001:2008 Certified International Journal of Engineering and Innovative Technology (IJEIT) Volume 3, Issue 1, July 2013
ISSN: 2277-375 Constructon of Trend Free Run Orders for Orthogonal rrays Usng Codes bstract: Sometmes when the expermental runs are carred out n a tme order sequence, the response can depend on the run
More informationThe Order Relation and Trace Inequalities for. Hermitian Operators
Internatonal Mathematcal Forum, Vol 3, 08, no, 507-57 HIKARI Ltd, wwwm-hkarcom https://doorg/0988/mf088055 The Order Relaton and Trace Inequaltes for Hermtan Operators Y Huang School of Informaton Scence
More informationOn the spectral norm of r-circulant matrices with the Pell and Pell-Lucas numbers
Türkmen and Gökbaş Journal of Inequaltes and Applcatons (06) 06:65 DOI 086/s3660-06-0997-0 R E S E A R C H Open Access On the spectral norm of r-crculant matrces wth the Pell and Pell-Lucas numbers Ramazan
More informationComparison of the Population Variance Estimators. of 2-Parameter Exponential Distribution Based on. Multiple Criteria Decision Making Method
Appled Mathematcal Scences, Vol. 7, 0, no. 47, 07-0 HIARI Ltd, www.m-hkar.com Comparson of the Populaton Varance Estmators of -Parameter Exponental Dstrbuton Based on Multple Crtera Decson Makng Method
More informationCOMPARISON OF SOME RELIABILITY CHARACTERISTICS BETWEEN REDUNDANT SYSTEMS REQUIRING SUPPORTING UNITS FOR THEIR OPERATIONS
Avalable onlne at http://sck.org J. Math. Comput. Sc. 3 (3), No., 6-3 ISSN: 97-537 COMPARISON OF SOME RELIABILITY CHARACTERISTICS BETWEEN REDUNDANT SYSTEMS REQUIRING SUPPORTING UNITS FOR THEIR OPERATIONS
More informationFormulas for the Determinant
page 224 224 CHAPTER 3 Determnants e t te t e 2t 38 A = e t 2te t e 2t e t te t 2e 2t 39 If 123 A = 345, 456 compute the matrx product A adj(a) What can you conclude about det(a)? For Problems 40 43, use
More informationThe Jacobsthal and Jacobsthal-Lucas Numbers via Square Roots of Matrices
Internatonal Mathematcal Forum, Vol 11, 2016, no 11, 513-520 HIKARI Ltd, wwwm-hkarcom http://dxdoorg/1012988/mf20166442 The Jacobsthal and Jacobsthal-Lucas Numbers va Square Roots of Matrces Saadet Arslan
More informationThe lower and upper bounds on Perron root of nonnegative irreducible matrices
Journal of Computatonal Appled Mathematcs 217 (2008) 259 267 wwwelsevercom/locate/cam The lower upper bounds on Perron root of nonnegatve rreducble matrces Guang-Xn Huang a,, Feng Yn b,keguo a a College
More informationLINEAR REGRESSION ANALYSIS. MODULE IX Lecture Multicollinearity
LINEAR REGRESSION ANALYSIS MODULE IX Lecture - 30 Multcollnearty Dr. Shalabh Department of Mathematcs and Statstcs Indan Insttute of Technology Kanpur 2 Remedes for multcollnearty Varous technques have
More informationNotes on Frequency Estimation in Data Streams
Notes on Frequency Estmaton n Data Streams In (one of) the data streamng model(s), the data s a sequence of arrvals a 1, a 2,..., a m of the form a j = (, v) where s the dentty of the tem and belongs to
More informationChapter 8 Indicator Variables
Chapter 8 Indcator Varables In general, e explanatory varables n any regresson analyss are assumed to be quanttatve n nature. For example, e varables lke temperature, dstance, age etc. are quanttatve n
More informationLinear Algebra and its Applications
Lnear Algebra and ts Applcatons 4 (00) 5 56 Contents lsts avalable at ScenceDrect Lnear Algebra and ts Applcatons journal homepage: wwwelsevercom/locate/laa Notes on Hlbert and Cauchy matrces Mroslav Fedler
More informationRandić Energy and Randić Estrada Index of a Graph
EUROPEAN JOURNAL OF PURE AND APPLIED MATHEMATICS Vol. 5, No., 202, 88-96 ISSN 307-5543 www.ejpam.com SPECIAL ISSUE FOR THE INTERNATIONAL CONFERENCE ON APPLIED ANALYSIS AND ALGEBRA 29 JUNE -02JULY 20, ISTANBUL
More informationA new eigenvalue inclusion set for tensors with its applications
COMPUTATIONAL SCIENCE RESEARCH ARTICLE A new egenvalue ncluson set for tensors wth ts applcatons Cal Sang 1 and Janxng Zhao 1 * Receved: 30 Deceber 2016 Accepted: 12 Aprl 2017 Frst Publshed: 20 Aprl 2017
More informationA new construction of 3-separable matrices via an improved decoding of Macula s construction
Dscrete Optmzaton 5 008 700 704 Contents lsts avalable at ScenceDrect Dscrete Optmzaton journal homepage: wwwelsevercom/locate/dsopt A new constructon of 3-separable matrces va an mproved decodng of Macula
More informationOn Graphs with Same Distance Distribution
Appled Mathematcs, 07, 8, 799-807 http://wwwscrporg/journal/am ISSN Onlne: 5-7393 ISSN Prnt: 5-7385 On Graphs wth Same Dstance Dstrbuton Xulang Qu, Xaofeng Guo,3 Chengy Unversty College, Jme Unversty,
More information2E Pattern Recognition Solutions to Introduction to Pattern Recognition, Chapter 2: Bayesian pattern classification
E395 - Pattern Recognton Solutons to Introducton to Pattern Recognton, Chapter : Bayesan pattern classfcaton Preface Ths document s a soluton manual for selected exercses from Introducton to Pattern Recognton
More informationVolume 18 Figure 1. Notation 1. Notation 2. Observation 1. Remark 1. Remark 2. Remark 3. Remark 4. Remark 5. Remark 6. Theorem A [2]. Theorem B [2].
Bulletn of Mathematcal Scences and Applcatons Submtted: 016-04-07 ISSN: 78-9634, Vol. 18, pp 1-10 Revsed: 016-09-08 do:10.1805/www.scpress.com/bmsa.18.1 Accepted: 016-10-13 017 ScPress Ltd., Swtzerland
More information4 Analysis of Variance (ANOVA) 5 ANOVA. 5.1 Introduction. 5.2 Fixed Effects ANOVA
4 Analyss of Varance (ANOVA) 5 ANOVA 51 Introducton ANOVA ANOVA s a way to estmate and test the means of multple populatons We wll start wth one-way ANOVA If the populatons ncluded n the study are selected
More informationChat eld, C. and A.J.Collins, Introduction to multivariate analysis. Chapman & Hall, 1980
MT07: Multvarate Statstcal Methods Mke Tso: emal mke.tso@manchester.ac.uk Webpage for notes: http://www.maths.manchester.ac.uk/~mkt/new_teachng.htm. Introducton to multvarate data. Books Chat eld, C. and
More informationDeterminants Containing Powers of Generalized Fibonacci Numbers
1 2 3 47 6 23 11 Journal of Integer Sequences, Vol 19 (2016), Artcle 1671 Determnants Contanng Powers of Generalzed Fbonacc Numbers Aram Tangboonduangjt and Thotsaporn Thanatpanonda Mahdol Unversty Internatonal
More informationThe L(2, 1)-Labeling on -Product of Graphs
Annals of Pure and Appled Mathematcs Vol 0, No, 05, 9-39 ISSN: 79-087X (P, 79-0888(onlne Publshed on 7 Aprl 05 wwwresearchmathscorg Annals of The L(, -Labelng on -Product of Graphs P Pradhan and Kamesh
More informationANSWERS. Problem 1. and the moment generating function (mgf) by. defined for any real t. Use this to show that E( U) var( U)
Econ 413 Exam 13 H ANSWERS Settet er nndelt 9 deloppgaver, A,B,C, som alle anbefales å telle lkt for å gøre det ltt lettere å stå. Svar er gtt . Unfortunately, there s a prntng error n the hnt of
More informationEstimation: Part 2. Chapter GREG estimation
Chapter 9 Estmaton: Part 2 9. GREG estmaton In Chapter 8, we have seen that the regresson estmator s an effcent estmator when there s a lnear relatonshp between y and x. In ths chapter, we generalzed the
More informationFundamental loop-current method using virtual voltage sources technique for special cases
Fundamental loop-current method usng vrtual voltage sources technque for specal cases George E. Chatzaraks, 1 Marna D. Tortorel 1 and Anastasos D. Tzolas 1 Electrcal and Electroncs Engneerng Departments,
More informationHowever, since P is a symmetric idempotent matrix, of P are either 0 or 1 [Eigen-values
Fall 007 Soluton to Mdterm Examnaton STAT 7 Dr. Goel. [0 ponts] For the general lnear model = X + ε, wth uncorrelated errors havng mean zero and varance σ, suppose that the desgn matrx X s not necessarly
More informationOn quasiperfect numbers
Notes on Number Theory and Dscrete Mathematcs Prnt ISSN 1310 5132, Onlne ISSN 2367 8275 Vol. 23, 2017, No. 3, 73 78 On quasperfect numbers V. Sva Rama Prasad 1 and C. Suntha 2 1 Nalla Malla Reddy Engneerng
More informationPARTIALLY BALANCED INCOMPLETE BLOCK DESIGNS
PARTIALLY BALANCED INCOMPLETE BLOCK DESIGNS V.K. Sharma I.A.S.R.I., Lbrary Avenue, New Delh-00. Introducton Balanced ncomplete block desgns, though have many optmal propertes, do not ft well to many expermental
More information8.4 COMPLEX VECTOR SPACES AND INNER PRODUCTS
SECTION 8.4 COMPLEX VECTOR SPACES AND INNER PRODUCTS 493 8.4 COMPLEX VECTOR SPACES AND INNER PRODUCTS All the vector spaces you have studed thus far n the text are real vector spaces because the scalars
More informationEstimating the Fundamental Matrix by Transforming Image Points in Projective Space 1
Estmatng the Fundamental Matrx by Transformng Image Ponts n Projectve Space 1 Zhengyou Zhang and Charles Loop Mcrosoft Research, One Mcrosoft Way, Redmond, WA 98052, USA E-mal: fzhang,cloopg@mcrosoft.com
More informationTHE Hadamard product of two nonnegative matrices and
IAENG Internatonal Journal of Appled Mathematcs 46:3 IJAM_46_3_5 Some New Bounds for the Hadamard Product of a Nonsngular M-matrx and Its Inverse Zhengge Huang Lgong Wang and Zhong Xu Abstract Some new
More informationLecture 3. Ax x i a i. i i
18.409 The Behavor of Algorthms n Practce 2/14/2 Lecturer: Dan Spelman Lecture 3 Scrbe: Arvnd Sankar 1 Largest sngular value In order to bound the condton number, we need an upper bound on the largest
More informationChapter 6. Supplemental Text Material
Chapter 6. Supplemental Text Materal S6-. actor Effect Estmates are Least Squares Estmates We have gven heurstc or ntutve explanatons of how the estmates of the factor effects are obtaned n the textboo.
More information2.3 Nilpotent endomorphisms
s a block dagonal matrx, wth A Mat dm U (C) In fact, we can assume that B = B 1 B k, wth B an ordered bass of U, and that A = [f U ] B, where f U : U U s the restrcton of f to U 40 23 Nlpotent endomorphsms
More informationSL n (F ) Equals its Own Derived Group
Internatonal Journal of Algebra, Vol. 2, 2008, no. 12, 585-594 SL n (F ) Equals ts Own Derved Group Jorge Macel BMCC-The Cty Unversty of New York, CUNY 199 Chambers street, New York, NY 10007, USA macel@cms.nyu.edu
More informationExistence of Two Conjugate Classes of A 5 within S 6. by Use of Character Table of S 6
Internatonal Mathematcal Forum, Vol. 8, 2013, no. 32, 1591-159 HIKARI Ltd, www.m-hkar.com http://dx.do.org/10.12988/mf.2013.3359 Exstence of Two Conjugate Classes of A 5 wthn S by Use of Character Table
More informationThe binomial transforms of the generalized (s, t )-Jacobsthal matrix sequence
Int. J. Adv. Appl. Math. and Mech. 6(3 (2019 14 20 (ISSN: 2347-2529 Journal homepage: www.jaamm.com IJAAMM Internatonal Journal of Advances n Appled Mathematcs and Mechancs The bnomal transforms of the
More informationDifference Equations
Dfference Equatons c Jan Vrbk 1 Bascs Suppose a sequence of numbers, say a 0,a 1,a,a 3,... s defned by a certan general relatonshp between, say, three consecutve values of the sequence, e.g. a + +3a +1
More informationOn a direct solver for linear least squares problems
ISSN 2066-6594 Ann. Acad. Rom. Sc. Ser. Math. Appl. Vol. 8, No. 2/2016 On a drect solver for lnear least squares problems Constantn Popa Abstract The Null Space (NS) algorthm s a drect solver for lnear
More informationSTAT 511 FINAL EXAM NAME Spring 2001
STAT 5 FINAL EXAM NAME Sprng Instructons: Ths s a closed book exam. No notes or books are allowed. ou may use a calculator but you are not allowed to store notes or formulas n the calculator. Please wrte
More informationDr. Shalabh Department of Mathematics and Statistics Indian Institute of Technology Kanpur
Analyss of Varance and Desgn of Experment-I MODULE VII LECTURE - 3 ANALYSIS OF COVARIANCE Dr Shalabh Department of Mathematcs and Statstcs Indan Insttute of Technology Kanpur Any scentfc experment s performed
More informationDr. Shalabh Department of Mathematics and Statistics Indian Institute of Technology Kanpur
Analyss of Varance and Desgn of Exerments-I MODULE III LECTURE - 2 EXPERIMENTAL DESIGN MODELS Dr. Shalabh Deartment of Mathematcs and Statstcs Indan Insttute of Technology Kanur 2 We consder the models
More informationComposite Hypotheses testing
Composte ypotheses testng In many hypothess testng problems there are many possble dstrbutons that can occur under each of the hypotheses. The output of the source s a set of parameters (ponts n a parameter
More informationOn locally nilpotent derivations of Boolean semirings
Daowsud et al., Cogent Mathematcs 2017, 4: 1351064 htts://do.org/10.1080/23311835.2017.1351064 PURE MATHEMATICS RESEARCH ARTICLE On locally nlotent dervatons of Boolean semrngs Katthaleeya Daowsud 1, Monrudee
More informationAn Introduction to Morita Theory
An Introducton to Morta Theory Matt Booth October 2015 Nov. 2017: made a few revsons. Thanks to Nng Shan for catchng a typo. My man reference for these notes was Chapter II of Bass s book Algebrac K-Theory
More informationON A DETERMINATION OF THE INITIAL FUNCTIONS FROM THE OBSERVED VALUES OF THE BOUNDARY FUNCTIONS FOR THE SECOND-ORDER HYPERBOLIC EQUATION
Advanced Mathematcal Models & Applcatons Vol.3, No.3, 2018, pp.215-222 ON A DETERMINATION OF THE INITIAL FUNCTIONS FROM THE OBSERVED VALUES OF THE BOUNDARY FUNCTIONS FOR THE SECOND-ORDER HYPERBOLIC EUATION
More informationMEM 255 Introduction to Control Systems Review: Basics of Linear Algebra
MEM 255 Introducton to Control Systems Revew: Bascs of Lnear Algebra Harry G. Kwatny Department of Mechancal Engneerng & Mechancs Drexel Unversty Outlne Vectors Matrces MATLAB Advanced Topcs Vectors A
More informationAsymptotics of the Solution of a Boundary Value. Problem for One-Characteristic Differential. Equation Degenerating into a Parabolic Equation
Nonl. Analyss and Dfferental Equatons, ol., 4, no., 5 - HIKARI Ltd, www.m-har.com http://dx.do.org/.988/nade.4.456 Asymptotcs of the Soluton of a Boundary alue Problem for One-Characterstc Dfferental Equaton
More informationLinear Approximation with Regularization and Moving Least Squares
Lnear Approxmaton wth Regularzaton and Movng Least Squares Igor Grešovn May 007 Revson 4.6 (Revson : March 004). 5 4 3 0.5 3 3.5 4 Contents: Lnear Fttng...4. Weghted Least Squares n Functon Approxmaton...
More informationAppendix for Causal Interaction in Factorial Experiments: Application to Conjoint Analysis
A Appendx for Causal Interacton n Factoral Experments: Applcaton to Conjont Analyss Mathematcal Appendx: Proofs of Theorems A. Lemmas Below, we descrbe all the lemmas, whch are used to prove the man theorems
More informationGraph Reconstruction by Permutations
Graph Reconstructon by Permutatons Perre Ille and Wllam Kocay* Insttut de Mathémathques de Lumny CNRS UMR 6206 163 avenue de Lumny, Case 907 13288 Marselle Cedex 9, France e-mal: lle@ml.unv-mrs.fr Computer
More informationA Local Variational Problem of Second Order for a Class of Optimal Control Problems with Nonsmooth Objective Function
A Local Varatonal Problem of Second Order for a Class of Optmal Control Problems wth Nonsmooth Objectve Functon Alexander P. Afanasev Insttute for Informaton Transmsson Problems, Russan Academy of Scences,
More informationA Solution of the Harry-Dym Equation Using Lattice-Boltzmannn and a Solitary Wave Methods
Appled Mathematcal Scences, Vol. 11, 2017, no. 52, 2579-2586 HIKARI Ltd, www.m-hkar.com https://do.org/10.12988/ams.2017.79280 A Soluton of the Harry-Dym Equaton Usng Lattce-Boltzmannn and a Soltary Wave
More informationSalmon: Lectures on partial differential equations. Consider the general linear, second-order PDE in the form. ,x 2
Salmon: Lectures on partal dfferental equatons 5. Classfcaton of second-order equatons There are general methods for classfyng hgher-order partal dfferental equatons. One s very general (applyng even to
More informationA CHARACTERIZATION OF ADDITIVE DERIVATIONS ON VON NEUMANN ALGEBRAS
Journal of Mathematcal Scences: Advances and Applcatons Volume 25, 2014, Pages 1-12 A CHARACTERIZATION OF ADDITIVE DERIVATIONS ON VON NEUMANN ALGEBRAS JIA JI, WEN ZHANG and XIAOFEI QI Department of Mathematcs
More information5 The Rational Canonical Form
5 The Ratonal Canoncal Form Here p s a monc rreducble factor of the mnmum polynomal m T and s not necessarly of degree one Let F p denote the feld constructed earler n the course, consstng of all matrces
More informationErratum: A Generalized Path Integral Control Approach to Reinforcement Learning
Journal of Machne Learnng Research 00-9 Submtted /0; Publshed 7/ Erratum: A Generalzed Path Integral Control Approach to Renforcement Learnng Evangelos ATheodorou Jonas Buchl Stefan Schaal Department of
More informationDepartment of Statistics University of Toronto STA305H1S / 1004 HS Design and Analysis of Experiments Term Test - Winter Solution
Department of Statstcs Unversty of Toronto STA35HS / HS Desgn and Analyss of Experments Term Test - Wnter - Soluton February, Last Name: Frst Name: Student Number: Instructons: Tme: hours. Ads: a non-programmable
More informationStatistical inference for generalized Pareto distribution based on progressive Type-II censored data with random removals
Internatonal Journal of Scentfc World, 2 1) 2014) 1-9 c Scence Publshng Corporaton www.scencepubco.com/ndex.php/ijsw do: 10.14419/jsw.v21.1780 Research Paper Statstcal nference for generalzed Pareto dstrbuton
More informationRegular product vague graphs and product vague line graphs
APPLIED & INTERDISCIPLINARY MATHEMATICS RESEARCH ARTICLE Regular product vague graphs and product vague lne graphs Ganesh Ghora 1 * and Madhumangal Pal 1 Receved: 26 December 2015 Accepted: 08 July 2016
More informationAPPENDIX A Some Linear Algebra
APPENDIX A Some Lnear Algebra The collecton of m, n matrces A.1 Matrces a 1,1,..., a 1,n A = a m,1,..., a m,n wth real elements a,j s denoted by R m,n. If n = 1 then A s called a column vector. Smlarly,
More informationAffine transformations and convexity
Affne transformatons and convexty The purpose of ths document s to prove some basc propertes of affne transformatons nvolvng convex sets. Here are a few onlne references for background nformaton: http://math.ucr.edu/
More informationApproximate D-optimal designs of experiments on the convex hull of a finite set of information matrices
Approxmate D-optmal desgns of experments on the convex hull of a fnte set of nformaton matrces Radoslav Harman, Mára Trnovská Department of Appled Mathematcs and Statstcs Faculty of Mathematcs, Physcs
More informationCSci 6974 and ECSE 6966 Math. Tech. for Vision, Graphics and Robotics Lecture 21, April 17, 2006 Estimating A Plane Homography
CSc 6974 and ECSE 6966 Math. Tech. for Vson, Graphcs and Robotcs Lecture 21, Aprl 17, 2006 Estmatng A Plane Homography Overvew We contnue wth a dscusson of the major ssues, usng estmaton of plane projectve
More informationPerron Vectors of an Irreducible Nonnegative Interval Matrix
Perron Vectors of an Irreducble Nonnegatve Interval Matrx Jr Rohn August 4 2005 Abstract As s well known an rreducble nonnegatve matrx possesses a unquely determned Perron vector. As the man result of
More informationConvexity preserving interpolation by splines of arbitrary degree
Computer Scence Journal of Moldova, vol.18, no.1(52), 2010 Convexty preservng nterpolaton by splnes of arbtrary degree Igor Verlan Abstract In the present paper an algorthm of C 2 nterpolaton of dscrete
More informationLINEAR REGRESSION ANALYSIS. MODULE IX Lecture Multicollinearity
LINEAR REGRESSION ANALYSIS MODULE IX Lecture - 31 Multcollnearty Dr. Shalabh Department of Mathematcs and Statstcs Indan Insttute of Technology Kanpur 6. Rdge regresson The OLSE s the best lnear unbased
More informationEcon107 Applied Econometrics Topic 3: Classical Model (Studenmund, Chapter 4)
I. Classcal Assumptons Econ7 Appled Econometrcs Topc 3: Classcal Model (Studenmund, Chapter 4) We have defned OLS and studed some algebrac propertes of OLS. In ths topc we wll study statstcal propertes
More informationA new Approach for Solving Linear Ordinary Differential Equations
, ISSN 974-57X (Onlne), ISSN 974-5718 (Prnt), Vol. ; Issue No. 1; Year 14, Copyrght 13-14 by CESER PUBLICATIONS A new Approach for Solvng Lnear Ordnary Dfferental Equatons Fawz Abdelwahd Department of
More informationSharp integral inequalities involving high-order partial derivatives. Journal Of Inequalities And Applications, 2008, v. 2008, article no.
Ttle Sharp ntegral nequaltes nvolvng hgh-order partal dervatves Authors Zhao, CJ; Cheung, WS Ctaton Journal Of Inequaltes And Applcatons, 008, v. 008, artcle no. 5747 Issued Date 008 URL http://hdl.handle.net/07/569
More informationHidden Markov Models & The Multivariate Gaussian (10/26/04)
CS281A/Stat241A: Statstcal Learnng Theory Hdden Markov Models & The Multvarate Gaussan (10/26/04) Lecturer: Mchael I. Jordan Scrbes: Jonathan W. Hu 1 Hdden Markov Models As a bref revew, hdden Markov models
More informationarxiv: v2 [quant-ph] 29 Jun 2018
Herarchy of Spn Operators, Quantum Gates, Entanglement, Tensor Product and Egenvalues Wll-Hans Steeb and Yorck Hardy arxv:59.7955v [quant-ph] 9 Jun 8 Internatonal School for Scentfc Computng, Unversty
More information2 More examples with details
Physcs 129b Lecture 3 Caltech, 01/15/19 2 More examples wth detals 2.3 The permutaton group n = 4 S 4 contans 4! = 24 elements. One s the dentty e. Sx of them are exchange of two objects (, j) ( to j and
More informationA Note on \Modules, Comodules, and Cotensor Products over Frobenius Algebras"
Chn. Ann. Math. 27B(4), 2006, 419{424 DOI: 10.1007/s11401-005-0025-z Chnese Annals of Mathematcs, Seres B c The Edtoral Oce of CAM and Sprnger-Verlag Berln Hedelberg 2006 A Note on \Modules, Comodules,
More informationAnti-van der Waerden numbers of 3-term arithmetic progressions.
Ant-van der Waerden numbers of 3-term arthmetc progressons. Zhanar Berkkyzy, Alex Schulte, and Mchael Young Aprl 24, 2016 Abstract The ant-van der Waerden number, denoted by aw([n], k), s the smallest
More informationA Note on Bound for Jensen-Shannon Divergence by Jeffreys
OPEN ACCESS Conference Proceedngs Paper Entropy www.scforum.net/conference/ecea- A Note on Bound for Jensen-Shannon Dvergence by Jeffreys Takuya Yamano, * Department of Mathematcs and Physcs, Faculty of
More informationRemarks on the Properties of a Quasi-Fibonacci-like Polynomial Sequence
Remarks on the Propertes of a Quas-Fbonacc-lke Polynomal Sequence Brce Merwne LIU Brooklyn Ilan Wenschelbaum Wesleyan Unversty Abstract Consder the Quas-Fbonacc-lke Polynomal Sequence gven by F 0 = 1,
More informationPsychology 282 Lecture #24 Outline Regression Diagnostics: Outliers
Psychology 282 Lecture #24 Outlne Regresson Dagnostcs: Outlers In an earler lecture we studed the statstcal assumptons underlyng the regresson model, ncludng the followng ponts: Formal statement of assumptons.
More information12 MATH 101A: ALGEBRA I, PART C: MULTILINEAR ALGEBRA. 4. Tensor product
12 MATH 101A: ALGEBRA I, PART C: MULTILINEAR ALGEBRA Here s an outlne of what I dd: (1) categorcal defnton (2) constructon (3) lst of basc propertes (4) dstrbutve property (5) rght exactness (6) localzaton
More informationDr. Shalabh Department of Mathematics and Statistics Indian Institute of Technology Kanpur
Analyss of Varance and Desgn of Experments- MODULE LECTURE - 6 EXPERMENTAL DESGN MODELS Dr. Shalabh Department of Mathematcs and Statstcs ndan nsttute of Technology Kanpur Two-way classfcaton wth nteractons
More informationErbakan University, Konya, Turkey. b Department of Mathematics, Akdeniz University, Antalya, Turkey. Published online: 28 Nov 2013.
Ths artcle was downloaded by: [Necmettn Erbaan Unversty] On: 24 March 2015, At: 05:44 Publsher: Taylor & Francs Informa Ltd Regstered n England and Wales Regstered Number: 1072954 Regstered offce: Mortmer
More informationThe Minimum Universal Cost Flow in an Infeasible Flow Network
Journal of Scences, Islamc Republc of Iran 17(2): 175-180 (2006) Unversty of Tehran, ISSN 1016-1104 http://jscencesutacr The Mnmum Unversal Cost Flow n an Infeasble Flow Network H Saleh Fathabad * M Bagheran
More informationFoundations of Arithmetic
Foundatons of Arthmetc Notaton We shall denote the sum and product of numbers n the usual notaton as a 2 + a 2 + a 3 + + a = a, a 1 a 2 a 3 a = a The notaton a b means a dvdes b,.e. ac = b where c s an
More information7. Products and matrix elements
7. Products and matrx elements 1 7. Products and matrx elements Based on the propertes of group representatons, a number of useful results can be derved. Consder a vector space V wth an nner product ψ
More informationChapter 13: Multiple Regression
Chapter 13: Multple Regresson 13.1 Developng the multple-regresson Model The general model can be descrbed as: It smplfes for two ndependent varables: The sample ft parameter b 0, b 1, and b are used to
More informationMore metrics on cartesian products
More metrcs on cartesan products If (X, d ) are metrc spaces for 1 n, then n Secton II4 of the lecture notes we defned three metrcs on X whose underlyng topologes are the product topology The purpose of
More informationLecture 10 Support Vector Machines II
Lecture 10 Support Vector Machnes II 22 February 2016 Taylor B. Arnold Yale Statstcs STAT 365/665 1/28 Notes: Problem 3 s posted and due ths upcomng Frday There was an early bug n the fake-test data; fxed
More informationApproximate Smallest Enclosing Balls
Chapter 5 Approxmate Smallest Enclosng Balls 5. Boundng Volumes A boundng volume for a set S R d s a superset of S wth a smple shape, for example a box, a ball, or an ellpsod. Fgure 5.: Boundng boxes Q(P
More informationChapter 11: Simple Linear Regression and Correlation
Chapter 11: Smple Lnear Regresson and Correlaton 11-1 Emprcal Models 11-2 Smple Lnear Regresson 11-3 Propertes of the Least Squares Estmators 11-4 Hypothess Test n Smple Lnear Regresson 11-4.1 Use of t-tests
More informationLECTURE 9 CANONICAL CORRELATION ANALYSIS
LECURE 9 CANONICAL CORRELAION ANALYSIS Introducton he concept of canoncal correlaton arses when we want to quantfy the assocatons between two sets of varables. For example, suppose that the frst set of
More informationSelf-complementing permutations of k-uniform hypergraphs
Dscrete Mathematcs Theoretcal Computer Scence DMTCS vol. 11:1, 2009, 117 124 Self-complementng permutatons of k-unform hypergraphs Artur Szymańsk A. Paweł Wojda Faculty of Appled Mathematcs, AGH Unversty
More informationCase Study of Markov Chains Ray-Knight Compactification
Internatonal Journal of Contemporary Mathematcal Scences Vol. 9, 24, no. 6, 753-76 HIKAI Ltd, www.m-har.com http://dx.do.org/.2988/cms.24.46 Case Study of Marov Chans ay-knght Compactfcaton HaXa Du and
More informationResearch Article Global Sufficient Optimality Conditions for a Special Cubic Minimization Problem
Mathematcal Problems n Engneerng Volume 2012, Artcle ID 871741, 16 pages do:10.1155/2012/871741 Research Artcle Global Suffcent Optmalty Condtons for a Specal Cubc Mnmzaton Problem Xaome Zhang, 1 Yanjun
More informationThe Pseudoblocks of Endomorphism Algebras
Internatonal Mathematcal Forum, 4, 009, no. 48, 363-368 The Pseudoblocks of Endomorphsm Algebras Ahmed A. Khammash Department of Mathematcal Scences, Umm Al-Qura Unversty P.O.Box 796, Makkah, Saud Araba
More informationModule 9. Lecture 6. Duality in Assignment Problems
Module 9 1 Lecture 6 Dualty n Assgnment Problems In ths lecture we attempt to answer few other mportant questons posed n earler lecture for (AP) and see how some of them can be explaned through the concept
More informationInexact Newton Methods for Inverse Eigenvalue Problems
Inexact Newton Methods for Inverse Egenvalue Problems Zheng-jan Ba Abstract In ths paper, we survey some of the latest development n usng nexact Newton-lke methods for solvng nverse egenvalue problems.
More informationProjective change between two Special (α, β)- Finsler Metrics
Internatonal Journal of Trend n Research and Development, Volume 2(6), ISSN 2394-9333 www.jtrd.com Projectve change between two Specal (, β)- Fnsler Metrcs Gayathr.K 1 and Narasmhamurthy.S.K 2 1 Assstant
More informationDistribution of subgraphs of random regular graphs
Dstrbuton of subgraphs of random regular graphs Zhcheng Gao Faculty of Busness Admnstraton Unversty of Macau Macau Chna zcgao@umac.mo N. C. Wormald Department of Combnatorcs and Optmzaton Unversty of Waterloo
More informationA New Evolutionary Computation Based Approach for Learning Bayesian Network
Avalable onlne at www.scencedrect.com Proceda Engneerng 15 (2011) 4026 4030 Advanced n Control Engneerng and Informaton Scence A New Evolutonary Computaton Based Approach for Learnng Bayesan Network Yungang
More informationA combinatorial proof of multiple angle formulas involving Fibonacci and Lucas numbers
Notes on Number Theory and Dscrete Mathematcs ISSN 1310 5132 Vol. 20, 2014, No. 5, 35 39 A combnatoral proof of multple angle formulas nvolvng Fbonacc and Lucas numbers Fernando Córes 1 and Dego Marques
More information