Chapter 2 OCDs in Completely Randomized Design Set-Up
|
|
- Darlene Booker
- 5 years ago
- Views:
Transcription
1 Chapter OCDs n Completely Randomzed Desgn Set-Up.1 Introducton We consder n ths chapter the one-way lnear model wth v treatments, c covarates and a total of n expermental unts. We work under the lnear model y = τ + c t=1 γ t z (t) + e, 1 j n, 1 v. (.1.1) where n (> 1) s the number of tmes the th treatment s replcated; clearly v n = n. (.1.) =1 For 1 j n,1 v, here y s the observaton arsng from the jth replcaton of the th treatment, τ effect due to the th treatment. In matrx notaton the above model can be represented as ) (Y, Xτ + Zγ, σ I n, (.1.3) where, Y s an observaton vector and X s the desgn matrx correspondng to vector of treatment effects τ v 1 and Z = ((z (t) )) s the desgn matrx correspondng to vector of covarate effects γ c 1 = (γ 1, γ,...,γ c ). Ths s referred to as one-way model wth covarates (wthout the general mean). Troya Lopes (198a, b) studed the nature of optmal allocaton of treatments and covarates n the above set-up for smultaneous estmaton of the (fxed) treatment effects (n the absence of the general effect) and the covarate effects wth maxmum effcency n the sense of mnmum generalzed varance. Ths s to note that the nformaton matrx wth respect to model (.1.3) s gven by σ I(η), where Sprnger Inda 015 P. Das et al., Optmal Covarate Desgns, DOI / _ 13
2 14 OCDs n Completely Randomzed Desgn Set-Up I(η) = ( X XX ) Z Z XZ Z (.1.4) and η = ( τ, γ ). The problem s to suggest an optmal allocaton scheme (for gven desgn parameters n, v,c) for effcent estmaton of the treatment effects as well as the covarate effects by ascertanng the values of the covarates for each one of them, assumng that each one s controllable and quanttatve wthn a stpulated fnte closed nterval. The nformaton matrx of γ s gven by σ I (γ) = Z Z Z X(X X) X Z (.1.5) where (X X) s a generalsed nverse of X X satsfyng X X ( X X ) X X = X X (cf. Rao 1973, p. 4). It s evdent that Z X(X X) X Z s a postve sem-defnte matrx. So from (.1.5), t follows that σ I (γ) Z Z (.1.6) n the Loewner order sense (vde Pukelshem 1993) where for two non-negatve defnte matrces A and B, A s sad to domnate B n the Lowener order sense f A B s a non-negatve defnte matrx. Equalty n (.1.6) s attaned whenever X Z = 0. (.1.7) If Z satsfes (.1.7), then treatment effects and covarate effects are orthogonally estmated. Agan under condton (.1.7), the nformaton matrx I(γ) reduces to I(γ) = Z Z.Thez-values are so chosen that Z Z s postve defnte so that from (.1.6) Var( γ t ) σ n v =1 j=1 z (t) σ n as z (t) [ 1, 1];, j, t. Now equalty n (.1.8) holds for all f and only f the Z-matrx s such that and (.1.8) z (s) z (t) = 0 s = t. (.1.9) z (t) =±1 (.1.10)
3 .1 Introducton 15 Condton (.1.7) mples that the estmators of ANOVA effects parameters or parametrc contrasts do not nterfere wth those of the covarate effects and condtons (.1.9) and (.1.10) mply that the estmators of each of the covarate effects are such that these are parwse uncorrelated, attanng the mnmum possble varance. Thus the covarate effects are estmated wth the maxmum effcency f and only f Z Z = ni c (.1.11) along wth (.1.7). The desgns allowng the estmators wth the mnmum varance are called globally optmal desgns (cf. Shah and Snha 1989, p. 143). Henceforth, we shall only be concerned wth such optmal estmaton of regresson parameters and by optmal covarate desgn, to be abbrevated as OCD hereafter, we shall only mean globally optmal desgn, unless otherwse mentoned. It s clear that condtons (.1.7) and (.1.11) hold smultaneously f and only f z s are necessarly +1or 1 and that condton (.1.7) s satsfed. It s dffcult to vsualze the Z-matrx satsfyng condtons (.1.7) and (.1.11).In the set-up of the model (.1.3), t transpres from Troya Lopes (198a) that optmal estmaton of the treatment effects and the covarates effects s possble when the treatment replcatons are all necessarly equal, assumng that n s a multple of v, the number of treatments. We set n = bv, where b s the common replcaton of treatments, henceforth. Das et al. (003) had represented each column of the Z- matrx by a v b matrx W wth elements of ±1, where the rows of W correspond to the v treatments and the columns of W correspond to dfferent replcaton numbers. Condton (.1.7) mples that the sum of each row of W should vansh. Agan, condton (.1.11) mples that the sum of products of the correspondng elements,.e. the Hadamard product of W (s) and W (t), defned n (.1.13) should also vansh, 1 s < t c. The above two facts can be represented n the followng schematc forms through the row totals and Hadamard product. Row Totals: Tr. Repl. no. Row 1... b Totals 1 0 W (s) = 0 (.1.1). (±1). v 0 Hadamard product of W (s) and W (t) (cf. Rao 1973, p. 30): W (s) W (t) = Tr. Repl. no b 1. v (w (s) w (t) ) (.1.13)
4 16 OCDs n Completely Randomzed Desgn Set-Up where * denotes Hadamard product. For orthogonalty of sth and tth columns of v b Z, t s requred that w (s) w (t) = 0. =1 j=1 The schematc representaton (.1.1), (.1.13) of Das et al. (003) s a breakthrough n the sense that handlng of Z-matrx has been made much easer and t has been followed throughout the monograph. Troya Lopes (198a) frst studed the nature of optmal allocaton of treatments and covarates n the above set-up when n v s an nteger. It may be noted that whenever condton (.1.7) s ensured, presence of the covarates n model (.1.3) does not pose any threat to the usual optmal treatment allocaton problem. In Sect.., followng Troya Lopes, we ntend to dscuss about the avalablty of Z-matrces satsfyng (.1.7) and (.1.11) when the treatment allocaton matrx X corresponds to equal allocaton number,.e. n stuatons where n s a multple of v. We wll wrte n = vb so that b s the common allocaton number of the v treatments under nvestgaton. The stuatons where (.1.7), (.1.11) and b = n v = nteger are satsfed, are dentfed as regular cases. Otherwse t s called a non-regular case. If the stuaton s nonregular, then t s not possble to allocate smultaneously the treatments and covarates optmally. For non-regular stuaton, effcent allocaton of treatments and covarates smultaneously can be done by usng other specfc optmalty crtera. Dey and Mukerjee (006) and Dutta et al. (014) consdered ths problem n non-regular stuatons and found D-optmal desgns n ths context. Detals are presented n Sect..3. It has been seen that Hadamard matrx plays a key role for constructng OCDs. Defnton of Hadamard matrx (cf. Hedayat et al. 1999, p. 145) s gven below: Defnton.1.1 A Hadamard matrx H t of order t s a t t matrx wth elements ±1 satsfyng H t H t = ti t.. Covarate Desgns Under Regular Cases Consder the case when n s a multple of v, that s n = vb where b s such that H b, Hadamard matrx of order b, exsts. We shall also consder some cases where b s even. Then ANOVA parameters as well as the covarate effect-parameters can be estmated orthgonally and/or most effcently. Ths holds smultaneously for c covarates and one can deduce maxmum possble value of c for ths to happen. As already mentoned, the most effcent estmaton of γ-components s possble when (.1.7) and (.1.11) are smultaneously satsfed and these condtons reduce, n terms of W-matrces defned n above, to C 1, C where
5 . Covarate Desgns Under Regular Cases 17 C 1. Each of the c W-matrces has all row-sums equal to zero; C. The grand total of all the entres n the Hadamard product of any two dstnct W-matrces reduces to zero. Now we defne optmum W-matrces for covarate desgns n CRD set-up. (..1) Defnton..1 Wth respect to model (.1.3), the c W-matrces correspondng to the c covarates are sad to be optmum f they satsfy condtons C 1 and C of (..1). In ths context, the followng results were deduced n Troya Lopes (198a). Theorem..1 Let c be the maxmum number of covarates that can be optmally accommodated. Then a lower bound to c s gven by (a) b 1 when v = odd, H b exsts; (b) (b 1) when v (mod4),h b exsts; (c) 4(b 1) when v 0(mod4),H b exsts; (d) 3v when b 0(mod4),H v exsts; (e) v when b (mod4),h v exsts. Proof Hadamard matrx H b s gven to exst and we wrte t as H b = (h 1, h,...,h b 1, 1). (..) The choce of optmum W-matrces s ndcated below one by one. The verfcaton of (..1) s mmedate and we leave t to the reader. The Kronecker product of two matrces s formally defned n Chap. 5 (Defnton 5.1.1) and t s used n the constructons below. (a) W ( j) v b = 1 v h j, 1 j b 1; (..3) (b) W ( j) v b = (1, 1) 1 v h j, 1 j b 1; } W (b 1+ j) v b = (1, 1) 1 v h j, 1 j b 1. (..4) (c) W ( j) v b = (1, 1, 1, 1) 1 v 4 h j, 1 j b 1; W (b 1+ j) v b = (1, 1, 1, 1) 1 v 4 h j, 1 j b 1; W ((b 1)+ j) v b = (1, 1, 1, 1) 1 v 4 h j, 1 j b 1; W (3(b 1)+ j) v b = (1, 1, 1, 1) 1 v 4 h j, 1 j b 1. (..5) (d) Let us represent a Hadamard matrx H v of order v as H v = ( h 1, h,...,h v). (..6)
6 18 OCDs n Completely Randomzed Desgn Set-Up W ( j) v b = (1, 1, 1, 1) 1 b h j, 1 j v; 4 W (v+ j) v b = (1, 1, 1, 1) 1 b h j, 1 j v; (..7) 4 W (v+ j) v b = (1, 1, 1, 1) 1 b h j, 1 j v. 4 (e) W ( j) v b = (1, 1) 1 b h j, 1 j v. (..8) Remark..1 In case (c), we can assume exstence of H v for all practcal purposes as v 0 (mod 4). So n ths case, an optmal desgn for maxmum possble v(b 1) optmum W-matrces can easly be constructed as W ((b 1)( 1)+ j) = h h j, = 1,,...,v, j = 1,,...,b 1. (..9) Ths was obtaned n (Rao et al. 003) where t was observed that OCDs n CRD and RBD have one to one correspondences wth mxed orthogonal array (MOA) (defnton gven n Chap. 3). Ths fact wll be dscussed n Sect. 3.3 of Chap. 3 n some further detals..3 Covarate Desgns Under Non-regular Cases Now we examne the stuatons where at least any one of the condtons (.1.7), (.1.11) and b = n v = nteger s volated. In that case, t s not possble to estmate smultaneously ANOVA parameters and γ-parameters orthogonally and/or most effcently. Thus we consder D-optmalty crteron to gve an effcent allocaton of treatments and covarates n Set-up (.1.1). Dey and Mukerjee (006) and Dutta et al. (014) have consdered ths stuaton and found D-optmal desgn. Here we dscuss ther contrbutons n ths drecton n detals. The vector of parameters θ, where θ = (μ 1, μ,...,μ v, γ 1,...,γ c ) (.3.1) s assumed to be estmable. The nformaton matrx for θ s gven by σ I(θ), where I(θ) = ( ) N T T Z, (.3.) Z N = Dag(n 1, n,...,n v ), (.3.3) T = (T 1, T,...,T v ), T = 1 n Z, (.3.4)
7 .3 Covarate Desgns Under Non-regular Cases 19 Z n c = (Z 1, Z,...,Z v ) (.3.5) and Z n c = z (1) 1 z (1) z() 1... z(c) 1 z()... z(c).. z (1) n z () n z (c) n. (.3.6) For D-optmalty, we have to maxmze the determnant of I(θ), denoted as det(i(θ)), wth respect to the desgn varables {z (t) } satsfyng z(t) [ 1, 1], 1 j n, 1 v and n s satsfyng (.1.). From (.3.) t s easy to see that ( v ) det(i(θ)) = n det(z Z T N 1 T) =1 ( v ) = n det(z Z =1 = det(n)det(c), n 1 T T ) (.3.7) where C = Z Z n 1 T T. (.3.8) Note that C s the nformaton matrx for the regresson coeffcents γ 1, γ,...,γ c. The maxmzaton of det(i(θ)) s done n two stages. In the frst stage, the maxmzaton s done for varyng z-values for fxed n s. Ths leads to an upper bound for det(i(θ)) obtaned through completely symmetrc C-matrces. At the second stage, maxmzaton s done for varyng n s subject to n = n, and ths leads to a suffcently small class N of contendng n = (n 1, n,...,n v ) s wheren the overall upper bound to det(i(θ)) belongs..3.1 Frst Stage of Maxmzaton Maxmsaton of I(θ) wth respect to z (t) [ 1, 1] s based on the followng lemma. Lemma.3.1 A necessary condton for maxmzaton of det(c) of(.3.8) wth respect to z (t) [ 1, 1], for fxed n s s that z (t) = ±1, j and t.
8 0 OCDs n Completely Randomzed Desgn Set-Up Proof From (.3.8), C can be expressed as where C = Z MZ = Z Z (.3.9) Z = MZ, M = dag(m 1, M,...,M v ), M = (I n n 1 1 n 1 n ). (.3.10) It s known that (cf. Gall and Kefer 1980; Wojtas 1964), det(z Z )smaxmum at the extreme entres of Z.Agan,asz (t) s are lnear n z (t) s, the determnant s maxmum at the extreme values of z (t) s for all, j and t. Hence the lemma follows. Theorem.3.1 For fxed {n } s satsfyng (.1.), ( v ) det(i(θ)) n {a + (c 1)b}(a b) c 1 (.3.11) =1 where a = n δ, b = ξ δ (.3.1) δ = v =1 n 1 δ, δ = 1(0) f n = odd(even) (.3.13) δ f both of n, δ are odd or even ξ = ξ(n, δ) = δ + 1 fn = odd, δ = even or n=even, δ = odd (.3.14) δ =greatest nteger less than equal to δ. Proof Because of Lemma.3.1, we restrct z (t) to the class χ ={z (t) : z (t) =±1}. From the Eq. (.3.8), we note that, c t,t,the(t, t ) th element of the C-matrx s gven by c t,t = { j z (t) ) z(t j z (t) j n z (t ) }, 1 t, t c. (.3.15) It follows from Wojtas (1964) that det(c) s maxmum when C s completely symmetrc wth all the dagonal elements equal to a and all off-dagonal elements equal to b where a and b are gven by max c tt and mn c 1 t c 1 t =t tt respectvely. Agan as c
9 .3 Covarate Desgns Under Non-regular Cases 1 z (t) =±1, j and t, for fxed n s, t can be deduced that max c tt = n δ = a, 1 t c mn c 1 t =t tt = ξ δ =b (.3.16) c where δ and ξ are gven n (.3.13) and (.3.14) respectvely. Therefore the theorem follows..3. Second Stage of Maxmzaton In vew of Theorem.3.1, we now consder the problem of maxmzng ( v ) g(n) = g(n 1, n,...,n v ) = n {a + (c 1)b}(a b) c 1 (.3.17) wth respect to n s subject to =1 v n = n, where a and b are gven by (.3.1) =1 (.3.14), so as to fnd the overall upper bound of det(i(θ)). The followng lemma helps to reduce the class N of n s where n = (n 1, n,...,n v ), satsfyng (.1.), to a subclass n whch maxmum of g(n) les. Lemma.3. Let n = ( n 1, n,...,n v) be a maxmzer of g(n) of (.3.17) subject to the condton (.1.). Then n cannot have () two unequal odd elements; () two even elements that dffer by more than ; () an even and an odd element that dffer by more than 1. Proof () Wthout loss of generalty t s assumed that n 1 and n be odd and n 1 n. Defne ñ = (ñ 1, ñ,...,ñ v ), where ñ 1 = n 1 + 1, ñ = n 1 and ñ = n = 1,. Note that ñ s satsfy condton (.1.). Then by Eq. (.3.17), g(ñ) g(n ) = ( v ) ( v ñ / ) ( ) {ã+(c 1) b}(ã b) c 1 {a +(c 1)b }(a b ) c 1 n =1 =1 = (n 1 +1)(n 1) {ã+(c 1) b}(ã b) c 1 n, 1 n {a +(c 1)b }(a b ) c 1 (.3.18) where, ã = n v =1 ñ 1 δ = n v =1 n 1 δ + 1 n n = a + 1 n n (.3.19)
10 OCDs n Completely Randomzed Desgn Set-Up Agan, v b = ξ =1 ñ 1 v δ ξ =1 ( 1 n 1 δ + n + 1 ) ( 1 1 n b + n + 1 ) 1 n. (.3.0) We consder the two cases b b and b > b separately. (a) Let b b. Then, as by (.3.19), ã > a,tfollowsthatg(ñ) >g(n ), whch s mpossble. (b) Let b > b and let b assume the hghest possble value gven n (.3.0). Then from (.3.18) (.3.0), t s seen that g(ñ) g(n ) > (n 1 + 1)(n 1) n > 1 (.3.1) 1 n whch s agan a contradcton. As the nequalty (.3.1) s true for the hghest value of b, t wll be true for all values of b n [b, b + 1 n n ] as ã > a. () If possble, let n have two even elements, say n 1 < n whch dffer by more than. Then as n () above, we reach at a contradcton by ncreasng n 1 by two and decreasng n by two. () If possble, let n have an even element n 1 and an odd element n whch dffer by more than 1. Case A: Let n 1 > n. Satsfyng (.1.), defne ñ = (ñ 1, ñ,...,ñ v ), where ñ 1 = n 1, ñ = n + and ñ = n = 1,. Then by Eq. (.3.17), we have g(ñ) g(n ) = (n 1 )(n + ){ã + (c 1) b}(ã b) c 1 (n 1 n ){a + (c 1)b }(a b ) c 1 (.3.) where, ã = n ( ñ 1 δ = n ) ( 1 n 1 δ + n 1 ) ( 1 n + = a + n 1 ) n + (.3.3) ( 1 b = ξ δ (ξ δ ) + n 1 ) ( 1 n b + n 1 ) n. (.3.4) We consder two cases when b b and b > b.for b b,tfollows,from(.3.) that g(ñ) >g(n ) as ã > a.agan,for b > b, we assume ts hghest value vz. b + ( 1 n 1 n ) from (.3.4) and use t n (.3.). It s seen that g(ñ) >g(n ), whch obvously holds for all other values of b > b as ã > a. So we reach at a contradcton that n s a maxmzer of g(n).
11 .3 Covarate Desgns Under Non-regular Cases 3 Case B: Let n 1 < n (.e. n 1 n 3), then we have the followng two cases: (a) n s not the only odd element of n. (b) n s the only odd element of n. For (a), let n have another odd element n 3. Then by part () of ths lemma, n =n 3. Defne ñ = (ñ 1, ñ,...,ñ v ), where ñ 1 = n 1 +, ñ = ñ 3 = n 1 and ñ = n = 1,, 3. Then by (.3.17) g(ñ) g(n ) = (n 1 + )(n 1) {ã + (c 1) b}(ã b) c 1 (n 1 n n 3 ) {a + (c 1)b }(a b. (.3.5) ) c 1 where, ã = n ( ñ 1 δ = n ) n 1 δ + n = a + n (.3.6) b = ξ δ (ξ δ ) + n b + n. (.3.7) If b b, then from (.3.5) and (.3.6) g(ñ) >g(n ) whch s a contradcton. If b > b, the above contradcton also holds by the same reasons as gven n Case A. For (b), let us defne ñ = (ñ 1, ñ,...,ñ v ) satsfyng (1.) where ñ 1 = n 1 +, ñ = n and ñ = n = 1,. Proceedng as before, t can be proved that ( 1 ã = a + n 1 ) ( n, b = 1 1 ) n. (.3.8) Usng (.3.8) n(.3.17), t s seen that g(ñ) g(n ) = (n 1 + )(n ) n 1 n ( n + c 1 ( n + c 1 c n ) c n Ths s agan a contradcton. Therefore the lemma follows. ) > 1 as(n n 1 ) 3. From Lemma.3., we get the followng theorem whose proof s mmedate. Theorem.3. Let o be an odd nteger, where o = n v or n v +1 accordng as n v s odd or even and n = ( n 1, n v),...,n be a maxmzer of g(n) of (.3.17) subject to n = n. Then n {o 1, o, o + 1}.
12 4 OCDs n Completely Randomzed Desgn Set-Up Lemma.3.3 If f, f and f + be the frequences of o, o 1 and o+1 respectvely, then the followng relatons f + f + f + = v; of + (o 1) f + (o + 1) f + = n, (.3.9) mnmze consderably the search for optmum n, forwhchg(n) s a maxmum. Let N ( N ) denote the class of n s satsfyng Theorem.3. and Lemma.3.. Remark.3.1 For gven n, vand c, let g(n ) be the maxmum of g(n) of (.3.17) over n = (n 1, n,...,n v ) subject to n = n. Then by Theorem.3.1 det(i(θ) g(n ). (.3.30) If a choce of {z (t) } exsts correspondng to n, such that equalty n (.3.30) holds, then n together wth {z (t) } gves a D-optmal desgn. Remark.3. If all n s are even, so that all the T s of (.3.4) may be made equal to zero, then t s possble to estmate the regresson parameters γ s orthogonally to the μ s. In that case, γ s are estmated most effcently wth the mnmum possble varance when Z Z = ni c. Remark.3.3 If n = n v = an even nteger for all, the stuaton reduces to regular case and then Remark.3.1 s n full agreement wth Troya Lopes (198a) and n that case γ s can be estmated most effcently so that each estmator has mnmum possble varance when Z Z = ni c. Remark.3.4 If the v levels of the sngle factor set-up are assumed to be the v level combnatons of m factors F 1,...,F m havng s 1,...,s m levels, respectvely (v = s ), then the optmum desgn for the sngle factor set-up s also optmum for the estmaton of γ and all effects up to m-factor nteractons whch can be obtaned through an orthogonal transformaton of γ and the mean vector μ correspondng to the v level combnatons..3.3 Examples Now we consder followng examples to llustrate the above method. Example.3.1 Let us consder the one-way set-up wth n = 1, v = 4. It follows that N ={(3, 3, 3, 3), (, 3, 3, 4), (,, 4, 4)} {(3 4 ), (, 3, 4), (, 4 )}.
13 .3 Covarate Desgns Under Non-regular Cases 5 (a) For c = 1, n = (3 4 ) s the unque maxmzer of g(n) and ths n together wth Z 1 = (1, 1, 1), Z = (1, 1, 1), Z 3 = (1, 1, 1), Z 4 = (1, 1, 1) gves a D-optmal desgn. (b) For c = both n = (, 3, 4) and (, 4 ) are maxmzers of g(n). () n = (, 4 ) and ( ) ( ) Z 1 =, Z 1 1 =, Z = , Z 4 = , gve a D-optmal desgn. () n = (, 3, 4) and ( ) Z 1 =, Z 1 1 = 1 1, Z 3 = 1 1, Z 4 = , also gve a D-optmal desgn. (c) For c = 3, 4, n = (, 4 ) s the unque maxmzer of g(n). Example.3. In one-way set-up wth n = 9, v= 3, c = 3, D-optmal desgn should be searched wthn the set {(, 3, 4), (3 3 )} of n. It s seen that for n = (3 3 ) and D 1 : Z (1) = 1 1 1, Z () = and Z (3) = 1 1 1, C = dag(8, 8, 8) and g(n) = Butforn = (, 3, 4), and ( ) D : Z (1) =, Z () = and Z (3) = It can be seen that C = 8I J 3, where J 3 s a 3 3 matrx contanng elements one only. Also g(, 3, 4) whch s equal to 15360, attans the upper bound n (.3.14) and g(, 3, 4) >g(3, 3, 3) mplyng that D s D-optmal. Agan for n = 9,v = 3, c = 4, t s noted that n = (, 3, 4) together wth ( ) D 3 : Z (1) =, Z () = and Z (3) = maxmzes g(n) of (.3.17) and hence gves a D-optmal desgn.
14 6 OCDs n Completely Randomzed Desgn Set-Up Remark.3.5 It s seen from the examples that the choce of optmum n depends on the number of the covarates used apart from the number of cells v n the set-up. Agan t s noted from (.3.7) that det(i(θ)) depends on two factors vz. det(n)(= n ) and det(c). Determnant of N ncreases as the homogenety between the n s ncreases subject to n = n. On the other hand det(c) ncreases, apart from c, wth the largeness of a and the smallness of b, whch agan are acheved by ncluson of maxmum number of even n s closed to n v. The number of odd n s subject to n = n, n between the even ones wth proper homogenety, actually strkes a balance between det(n) and det(c). It s also seen that, when c s small, det(n) s the domnant factor, whle, f c s large det(c) becomes the domnant factor. Incdentally, the above analyss s based on the work n Dutta et al. (014) and t mproves over what was acheved n Dey and Mukerjee (006). References Das K, Mandal NK, Snha BK (003) Optmal expermental desgns wth covarates. J Stat Plan Inference 115:73 85 Dey A, Mukerjee R (006) D-optmal desgns for covarate models. Statstcs 40: Dutta G, Das P, Mandal NK (014) D-Optmal desgns for covarate models. Commun Stat Theory Methods 43: Gall Z, Kefer J (1980) D-optmum weghng desgns. Ann Stat 8: Hedayat AS, Sloane NJA, Stufken J (1999) Orthogonal arrays: theory and applcatons. Sprnger, New York Pukelshem F (1993) Optmal desgn of experments. Wley, New York Rao CR (1973) Lnear statstcal nference and ts applcatons. Wley, New York Rao PSSNVP, Rao SB, Saha GM, Snha BK (003) Optmal desgns for covarates models and mxed orthogonal arrays. Electron Notes Dscret Math 15: Shah KR, Snha BK (1989) Theory of optmal desgns. lecture notes n statstcs, Seres 54. Sprnger, New York Troya Lopes J (198a) Optmal desgns for covarate models. J Stat Plan Inference 6: Troya Lopes J (198b) Cyclc desgns for a covarate model. J Stat Plan Inference 7:49 75 Wojtas M (1964) On Hadamard s nequalty for the determnants of order non-dvsble by 4. Colloq Math 1:73 83
15
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 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 informationChapter 5. Solution of System of Linear Equations. Module No. 6. Solution of Inconsistent and Ill Conditioned Systems
Numercal Analyss by Dr. Anta Pal Assstant Professor Department of Mathematcs Natonal Insttute of Technology Durgapur Durgapur-713209 emal: anta.bue@gmal.com 1 . Chapter 5 Soluton of System of Lnear Equatons
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 informationThe Multiple Classical Linear Regression Model (CLRM): Specification and Assumptions. 1. Introduction
ECONOMICS 5* -- NOTE (Summary) ECON 5* -- NOTE The Multple Classcal Lnear Regresson Model (CLRM): Specfcaton and Assumptons. Introducton CLRM stands for the Classcal Lnear Regresson Model. The CLRM s also
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 informationChapter 7 Generalized and Weighted Least Squares Estimation. In this method, the deviation between the observed and expected values of
Chapter 7 Generalzed and Weghted Least Squares Estmaton The usual lnear regresson model assumes that all the random error components are dentcally and ndependently dstrbuted wth constant varance. When
More informationSolutions Homework 4 March 5, 2018
1 Solutons Homework 4 March 5, 018 Soluton to Exercse 5.1.8: Let a IR be a translaton and c > 0 be a re-scalng. ˆb1 (cx + a) cx n + a (cx 1 + a) c x n x 1 cˆb 1 (x), whch shows ˆb 1 s locaton nvarant and
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 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 informationNumber of cases Number of factors Number of covariates Number of levels of factor i. Value of the dependent variable for case k
ANOVA Model and Matrx Computatons Notaton The followng notaton s used throughout ths chapter unless otherwse stated: N F CN Y Z j w W Number of cases Number of factors Number of covarates Number of levels
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 informationFirst day August 1, Problems and Solutions
FOURTH INTERNATIONAL COMPETITION FOR UNIVERSITY STUDENTS IN MATHEMATICS July 30 August 4, 997, Plovdv, BULGARIA Frst day August, 997 Problems and Solutons Problem. Let {ε n } n= be a sequence of postve
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 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 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 information3.1 Expectation of Functions of Several Random Variables. )' be a k-dimensional discrete or continuous random vector, with joint PMF p (, E X E X1 E X
Statstcs 1: Probablty Theory II 37 3 EPECTATION OF SEVERAL RANDOM VARIABLES As n Probablty Theory I, the nterest n most stuatons les not on the actual dstrbuton of a random vector, but rather on a number
More informationDr. Shalabh Department of Mathematics and Statistics Indian Institute of Technology Kanpur
Analyss of Varance and Desgn of Experment-I MODULE VIII LECTURE - 34 ANALYSIS OF VARIANCE IN RANDOM-EFFECTS MODEL AND MIXED-EFFECTS EFFECTS MODEL Dr Shalabh Department of Mathematcs and Statstcs Indan
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 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 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 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 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 informationInner Product. Euclidean Space. Orthonormal Basis. Orthogonal
Inner Product Defnton 1 () A Eucldean space s a fnte-dmensonal vector space over the reals R, wth an nner product,. Defnton 2 (Inner Product) An nner product, on a real vector space X s a symmetrc, blnear,
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 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 informationA Comparative Study for Estimation Parameters in Panel Data Model
A Comparatve Study for Estmaton Parameters n Panel Data Model Ahmed H. Youssef and Mohamed R. Abonazel hs paper examnes the panel data models when the regresson coeffcents are fxed random and mxed and
More informationa b a In case b 0, a being divisible by b is the same as to say that
Secton 6.2 Dvsblty among the ntegers An nteger a ε s dvsble by b ε f there s an nteger c ε such that a = bc. Note that s dvsble by any nteger b, snce = b. On the other hand, a s dvsble by only f a = :
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 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 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 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 informationStanford University CS359G: Graph Partitioning and Expanders Handout 4 Luca Trevisan January 13, 2011
Stanford Unversty CS359G: Graph Parttonng and Expanders Handout 4 Luca Trevsan January 3, 0 Lecture 4 In whch we prove the dffcult drecton of Cheeger s nequalty. As n the past lectures, consder an undrected
More informationTHE CHINESE REMAINDER THEOREM. We should thank the Chinese for their wonderful remainder theorem. Glenn Stevens
THE CHINESE REMAINDER THEOREM KEITH CONRAD We should thank the Chnese for ther wonderful remander theorem. Glenn Stevens 1. Introducton The Chnese remander theorem says we can unquely solve any par of
More informationChapter 12 Analysis of Covariance
Chapter Analyss of Covarance Any scentfc experment s performed to know somethng that s unknown about a group of treatments and to test certan hypothess about the correspondng treatment effect When varablty
More informationprinceton univ. F 17 cos 521: Advanced Algorithm Design Lecture 7: LP Duality Lecturer: Matt Weinberg
prnceton unv. F 17 cos 521: Advanced Algorthm Desgn Lecture 7: LP Dualty Lecturer: Matt Wenberg Scrbe: LP Dualty s an extremely useful tool for analyzng structural propertes of lnear programs. Whle there
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 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 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 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 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 informationFirst Year Examination Department of Statistics, University of Florida
Frst Year Examnaton Department of Statstcs, Unversty of Florda May 7, 010, 8:00 am - 1:00 noon Instructons: 1. You have four hours to answer questons n ths examnaton.. You must show your work to receve
More informationn α j x j = 0 j=1 has a nontrivial solution. Here A is the n k matrix whose jth column is the vector for all t j=0
MODULE 2 Topcs: Lnear ndependence, bass and dmenson We have seen that f n a set of vectors one vector s a lnear combnaton of the remanng vectors n the set then the span of the set s unchanged f that vector
More informationFeb 14: Spatial analysis of data fields
Feb 4: Spatal analyss of data felds Mappng rregularly sampled data onto a regular grd Many analyss technques for geophyscal data requre the data be located at regular ntervals n space and/or tme. hs s
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 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 informationSystems of Equations (SUR, GMM, and 3SLS)
Lecture otes on Advanced Econometrcs Takash Yamano Fall Semester 4 Lecture 4: Sstems of Equatons (SUR, MM, and 3SLS) Seemngl Unrelated Regresson (SUR) Model Consder a set of lnear equatons: $ + ɛ $ + ɛ
More informationLectures - Week 4 Matrix norms, Conditioning, Vector Spaces, Linear Independence, Spanning sets and Basis, Null space and Range of a Matrix
Lectures - Week 4 Matrx norms, Condtonng, Vector Spaces, Lnear Independence, Spannng sets and Bass, Null space and Range of a Matrx Matrx Norms Now we turn to assocatng a number to each matrx. We could
More information763622S ADVANCED QUANTUM MECHANICS Solution Set 1 Spring c n a n. c n 2 = 1.
7636S ADVANCED QUANTUM MECHANICS Soluton Set 1 Sprng 013 1 Warm-up Show that the egenvalues of a Hermtan operator  are real and that the egenkets correspondng to dfferent egenvalues are orthogonal (b)
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 informationThe Second Anti-Mathima on Game Theory
The Second Ant-Mathma on Game Theory Ath. Kehagas December 1 2006 1 Introducton In ths note we wll examne the noton of game equlbrum for three types of games 1. 2-player 2-acton zero-sum games 2. 2-player
More informationNegative Binomial Regression
STATGRAPHICS Rev. 9/16/2013 Negatve Bnomal Regresson Summary... 1 Data Input... 3 Statstcal Model... 3 Analyss Summary... 4 Analyss Optons... 7 Plot of Ftted Model... 8 Observed Versus Predcted... 10 Predctons...
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 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 informationLINEAR REGRESSION ANALYSIS. MODULE VIII Lecture Indicator Variables
LINEAR REGRESSION ANALYSIS MODULE VIII Lecture - 7 Indcator Varables Dr. Shalabh Department of Maematcs and Statstcs Indan Insttute of Technology Kanpur Indcator varables versus quanttatve explanatory
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 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 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 informationU.C. Berkeley CS294: Spectral Methods and Expanders Handout 8 Luca Trevisan February 17, 2016
U.C. Berkeley CS94: Spectral Methods and Expanders Handout 8 Luca Trevsan February 7, 06 Lecture 8: Spectral Algorthms Wrap-up In whch we talk about even more generalzatons of Cheeger s nequaltes, and
More informationLecture 20: Lift and Project, SDP Duality. Today we will study the Lift and Project method. Then we will prove the SDP duality theorem.
prnceton u. sp 02 cos 598B: algorthms and complexty Lecture 20: Lft and Project, SDP Dualty Lecturer: Sanjeev Arora Scrbe:Yury Makarychev Today we wll study the Lft and Project method. Then we wll prove
More informationSingular Value Decomposition: Theory and Applications
Sngular Value Decomposton: Theory and Applcatons Danel Khashab Sprng 2015 Last Update: March 2, 2015 1 Introducton A = UDV where columns of U and V are orthonormal and matrx D s dagonal wth postve real
More informationChapter 9: Statistical Inference and the Relationship between Two Variables
Chapter 9: Statstcal Inference and the Relatonshp between Two Varables Key Words The Regresson Model The Sample Regresson Equaton The Pearson Correlaton Coeffcent Learnng Outcomes After studyng ths chapter,
More informationMATH 241B FUNCTIONAL ANALYSIS - NOTES EXAMPLES OF C ALGEBRAS
MATH 241B FUNCTIONAL ANALYSIS - NOTES EXAMPLES OF C ALGEBRAS These are nformal notes whch cover some of the materal whch s not n the course book. The man purpose s to gve a number of nontrval examples
More informationResource Allocation and Decision Analysis (ECON 8010) Spring 2014 Foundations of Regression Analysis
Resource Allocaton and Decson Analss (ECON 800) Sprng 04 Foundatons of Regresson Analss Readng: Regresson Analss (ECON 800 Coursepak, Page 3) Defntons and Concepts: Regresson Analss statstcal technques
More informationErrors for Linear Systems
Errors for Lnear Systems When we solve a lnear system Ax b we often do not know A and b exactly, but have only approxmatons  and ˆb avalable. Then the best thng we can do s to solve ˆx ˆb exactly whch
More informationPrimer on High-Order Moment Estimators
Prmer on Hgh-Order Moment Estmators Ton M. Whted July 2007 The Errors-n-Varables Model We wll start wth the classcal EIV for one msmeasured regressor. The general case s n Erckson and Whted Econometrc
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 informationModule 3 LOSSY IMAGE COMPRESSION SYSTEMS. Version 2 ECE IIT, Kharagpur
Module 3 LOSSY IMAGE COMPRESSION SYSTEMS Verson ECE IIT, Kharagpur Lesson 6 Theory of Quantzaton Verson ECE IIT, Kharagpur Instructonal Objectves At the end of ths lesson, the students should be able to:
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 informationBeyond Zudilin s Conjectured q-analog of Schmidt s problem
Beyond Zudln s Conectured q-analog of Schmdt s problem Thotsaporn Ae Thanatpanonda thotsaporn@gmalcom Mathematcs Subect Classfcaton: 11B65 33B99 Abstract Usng the methodology of (rgorous expermental mathematcs
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 informationarxiv: v1 [math.co] 1 Mar 2014
Unon-ntersectng set systems Gyula O.H. Katona and Dánel T. Nagy March 4, 014 arxv:1403.0088v1 [math.co] 1 Mar 014 Abstract Three ntersecton theorems are proved. Frst, we determne the sze of the largest
More informationUsing T.O.M to Estimate Parameter of distributions that have not Single Exponential Family
IOSR Journal of Mathematcs IOSR-JM) ISSN: 2278-5728. Volume 3, Issue 3 Sep-Oct. 202), PP 44-48 www.osrjournals.org Usng T.O.M to Estmate Parameter of dstrbutons that have not Sngle Exponental Famly Jubran
More informationSolutions to exam in SF1811 Optimization, Jan 14, 2015
Solutons to exam n SF8 Optmzaton, Jan 4, 25 3 3 O------O -4 \ / \ / The network: \/ where all lnks go from left to rght. /\ / \ / \ 6 O------O -5 2 4.(a) Let x = ( x 3, x 4, x 23, x 24 ) T, where the varable
More informationWelfare Properties of General Equilibrium. What can be said about optimality properties of resource allocation implied by general equilibrium?
APPLIED WELFARE ECONOMICS AND POLICY ANALYSIS Welfare Propertes of General Equlbrum What can be sad about optmalty propertes of resource allocaton mpled by general equlbrum? Any crteron used to compare
More informationLecture 2: Prelude to the big shrink
Lecture 2: Prelude to the bg shrnk Last tme A slght detour wth vsualzaton tools (hey, t was the frst day... why not start out wth somethng pretty to look at?) Then, we consdered a smple 120a-style regresson
More informationMin Cut, Fast Cut, Polynomial Identities
Randomzed Algorthms, Summer 016 Mn Cut, Fast Cut, Polynomal Identtes Instructor: Thomas Kesselhem and Kurt Mehlhorn 1 Mn Cuts n Graphs Lecture (5 pages) Throughout ths secton, G = (V, E) s a mult-graph.
More informationPerfect Competition and the Nash Bargaining Solution
Perfect Competton and the Nash Barganng Soluton Renhard John Department of Economcs Unversty of Bonn Adenauerallee 24-42 53113 Bonn, Germany emal: rohn@un-bonn.de May 2005 Abstract For a lnear exchange
More informationStructure and Drive Paul A. Jensen Copyright July 20, 2003
Structure and Drve Paul A. Jensen Copyrght July 20, 2003 A system s made up of several operatons wth flow passng between them. The structure of the system descrbes the flow paths from nputs to outputs.
More informationEdge Isoperimetric Inequalities
November 7, 2005 Ross M. Rchardson Edge Isopermetrc Inequaltes 1 Four Questons Recall that n the last lecture we looked at the problem of sopermetrc nequaltes n the hypercube, Q n. Our noton of boundary
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 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 informationMMA and GCMMA two methods for nonlinear optimization
MMA and GCMMA two methods for nonlnear optmzaton Krster Svanberg Optmzaton and Systems Theory, KTH, Stockholm, Sweden. krlle@math.kth.se Ths note descrbes the algorthms used n the author s 2007 mplementatons
More informationMath 217 Fall 2013 Homework 2 Solutions
Math 17 Fall 013 Homework Solutons Due Thursday Sept. 6, 013 5pm Ths homework conssts of 6 problems of 5 ponts each. The total s 30. You need to fully justfy your answer prove that your functon ndeed has
More informationLecture 5 Decoding Binary BCH Codes
Lecture 5 Decodng Bnary BCH Codes In ths class, we wll ntroduce dfferent methods for decodng BCH codes 51 Decodng the [15, 7, 5] 2 -BCH Code Consder the [15, 7, 5] 2 -code C we ntroduced n the last lecture
More informationThe internal structure of natural numbers and one method for the definition of large prime numbers
The nternal structure of natural numbers and one method for the defnton of large prme numbers Emmanul Manousos APM Insttute for the Advancement of Physcs and Mathematcs 3 Poulou str. 53 Athens Greece Abstract
More informatione - c o m p a n i o n
OPERATIONS RESEARCH http://dxdoorg/0287/opre007ec e - c o m p a n o n ONLY AVAILABLE IN ELECTRONIC FORM 202 INFORMS Electronc Companon Generalzed Quantty Competton for Multple Products and Loss of Effcency
More informationMaximum Likelihood Estimation of Binary Dependent Variables Models: Probit and Logit. 1. General Formulation of Binary Dependent Variables Models
ECO 452 -- OE 4: Probt and Logt Models ECO 452 -- OE 4 Maxmum Lkelhood Estmaton of Bnary Dependent Varables Models: Probt and Logt hs note demonstrates how to formulate bnary dependent varables models
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 informationCOS 521: Advanced Algorithms Game Theory and Linear Programming
COS 521: Advanced Algorthms Game Theory and Lnear Programmng Moses Charkar February 27, 2013 In these notes, we ntroduce some basc concepts n game theory and lnear programmng (LP). We show a connecton
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 informationMATH 281A: Homework #6
MATH 28A: Homework #6 Jongha Ryu Due date: November 8, 206 Problem. (Problem 2..2. Soluton. If X,..., X n Bern(p, then T = X s a complete suffcent statstc. Our target s g(p = p, and the nave guess suggested
More informationStatistics for Economics & Business
Statstcs for Economcs & Busness Smple Lnear Regresson Learnng Objectves In ths chapter, you learn: How to use regresson analyss to predct the value of a dependent varable based on an ndependent varable
More informationGoogle PageRank with Stochastic Matrix
Google PageRank wth Stochastc Matrx Md. Sharq, Puranjt Sanyal, Samk Mtra (M.Sc. Applcatons of Mathematcs) Dscrete Tme Markov Chan Let S be a countable set (usually S s a subset of Z or Z d or R or R d
More informationFACTORIZATION IN KRULL MONOIDS WITH INFINITE CLASS GROUP
C O L L O Q U I U M M A T H E M A T I C U M VOL. 80 1999 NO. 1 FACTORIZATION IN KRULL MONOIDS WITH INFINITE CLASS GROUP BY FLORIAN K A I N R A T H (GRAZ) Abstract. Let H be a Krull monod wth nfnte class
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 information10-801: Advanced Optimization and Randomized Methods Lecture 2: Convex functions (Jan 15, 2014)
0-80: Advanced Optmzaton and Randomzed Methods Lecture : Convex functons (Jan 5, 04) Lecturer: Suvrt Sra Addr: Carnege Mellon Unversty, Sprng 04 Scrbes: Avnava Dubey, Ahmed Hefny Dsclamer: These notes
More informationExample: (13320, 22140) =? Solution #1: The divisors of are 1, 2, 3, 4, 5, 6, 9, 10, 12, 15, 18, 20, 27, 30, 36, 41,
The greatest common dvsor of two ntegers a and b (not both zero) s the largest nteger whch s a common factor of both a and b. We denote ths number by gcd(a, b), or smply (a, b) when there s no confuson
More informationTHE WEIGHTED WEAK TYPE INEQUALITY FOR THE STRONG MAXIMAL FUNCTION
THE WEIGHTED WEAK TYPE INEQUALITY FO THE STONG MAXIMAL FUNCTION THEMIS MITSIS Abstract. We prove the natural Fefferman-Sten weak type nequalty for the strong maxmal functon n the plane, under the assumpton
More informationPolynomial Regression Models
LINEAR REGRESSION ANALYSIS MODULE XII Lecture - 6 Polynomal Regresson Models Dr. Shalabh Department of Mathematcs and Statstcs Indan Insttute of Technology Kanpur Test of sgnfcance To test the sgnfcance
More information