An epsilon-based measure of efficiency in DEA revisited -A third pole of technical efficiency-

Size: px
Start display at page:

Download "An epsilon-based measure of efficiency in DEA revisited -A third pole of technical efficiency-"

Transcription

1 GRIPS Plicy Ifrmati Ceter Discussi Paper : 09-2 A epsil-based measure f efficiecy i DEA revisited -A third ple f techical efficiecy- Karu Te Natial Graduate Istitute fr Plicy Studies Rppgi, Miat-ku, Tky , Japa Miki Tsutsui Cetral Research Istitute f Electric Pwer Idustry 2-- Iwad Kita, Kmae-shi, Tky 20-85, Japa Abstract I DEA, we have tw measures f techical efficiecy with differet characteristics: radial ad -radial. I this paper we cmpile them it a cmpsite mdel called epsil-based measure (EBM). Fr this purpse we itrduce tw parameters which cect radial ad -radial mdels. These tw parameters are btaied frm the ewly defied affiity ide betwee iputs r utputs alg with pricipal cmpet aalysis the affiity matri. Thus, EBM takes it accut diversity f iput/utput data ad their relative imprtace fr measurig techical efficiecy. Keywrds: Data evelpmet aalysis, Radial, N-radial, CCR, SBM, EBM, Pricipal cmpet aalysis. Itrducti DEA (data evelpmet aalysis) is a data drive tl fr measurig efficiecy f decisi makig uits (DMU) ad shws a sharp ctrast t s-called parametric methds such as SFA. The latter methds assume specific prducti fucti frms t be idetified. This assumpti is t s reasable i several istaces ad aspects. Sice DEA ca deal with multiple iput vs. multiple utput relatis i a sigle framewrk, it has bee becmig a methd f chice fr efficiecy evaluati i recet days. Hwever, DEA has several shrtcmigs t be eplred further. I DEA, we have tw measures f techical efficiecy with differet characteristics: radial ad -radial. Histrically, the radial measure, represeted by the CCR mdel (Chares, Cper ad Rhdes [5]), was the first DEA mdel, whereas the -radial mdel, represeted by the SBM mdel (slacks-based measure by Te [8], see als Cper et al. [6]) was a latecmer. Fr istace, i the iput-rieted case, the CCR deals maily with prprtiate reducti f iput resurces. I ther wrds, if the rgaisatial uit uder study, als kw as a DMU, has tw iputs, this mdel aims at btaiig the maimum rate f reducti with the same prprti, i.e. a radial ctracti i the tw iputs that ca prduce the curret utputs. I ctrast, the -radial mdels put aside

2 GRIPS Plicy Ifrmati Ceter Discussi Paper : 09-2 the assumpti f prprtiate ctracti i iputs ad aim at btaiig maimum rates f reducti i iputs that may discard varyig prprtis f rigial iput resurces. I this paper, after itrducig radial ad -radial mdels briefly, we prpse a cmpsite mdel which cmbies bth mdels i a uified framewrk. This mdel has tw parameters: e scalar ad e vectr. I rder t determie these tw parameters, we itrduce a ew affiity ide assciated with iputs r utputs. We apply pricipal cmpet aalysis t thus defied affiity matri. This paper uflds as fllws. I Secti 2, we briefly survey radial ad -radial mdels i DEA. I Secti 3, we prpse the epsil-based measure f efficiecy (EBM). EBM eeds tw parameters. After bservig tw etreme diversities f dataset, we itrduce a ew crrelati cefficiet called affiity ide i Secti 4. We utilize this ide fr defiig affiity matri amg iput/utput data. Frm this matri we derive tw parameters fr EBM i Secti 5. We discuss ratiality f the scheme i Secti 6. We demstrate illustrative eamples i Secti 7. I Secti 8 we eted the mdel t ther rietatis ad variable returs-t-scale evirmet. We cclude this paper i Secti Radial ad -radial measures f efficiecy I this secti we itrduce the CCR ad SBM mdels as represetative radial ad -radial measures f efficiecy respectively, ad pit ut their shrtcmigs. Thrughut this paper, we deal with DMUs ( =, K, ) havig m iputs ( i =, K, m) ad s utputs ( r =, K, s). The iput m ad utput matrices are deted by X = { } R s ad Y = { y } R X >0 ad Y>0. 2. The CCR ad SBM Mdels We briefly eplai the CCR ad SBM mdels, ad cmpare their iefficiecy status. (a) The CCR Mdel i r, respectively. We assume The iput-rieted CCR mdel evaluates the techical efficiecy the fllwig liear prgram: θ f DMU (, y ) by slvig [CCR-I] θ = mi θ θ, λ, s () subect t 2

3 GRIPS Plicy Ifrmati Ceter Discussi Paper : 09-2 θ = Xλ + s y Yλ λ 0s, 0, where λ represets the itesity vectr ad s detes the -radial slacks. Usually, we slve [CCR-I] i a tw phase prcess. I the first phase, we slve [CCR-I] ad btai θ (weak efficiecy). The, i the secd phase, we maimize m si i (2) i terms f λ ad s, subect t (2) ad θ = θ. (b) The SBM Mdel Here, we ctiue with iput rietati csistet with ur epsiti f the CCR mdel i the precedig paragraph. The iput-rieted SBM mdel uder the cstat returs-t-scale assumpti evaluates the efficiecy τ f DMU (, y) by slvig the fllwig liear prgram where the abbreviatis I ad C idicate Iput-rieted ad Cstat-returs-t-scale, respectively. [SBM-I-C] τ = mi m subect t m s i i i iλ i = = + s ( i =, K, m) i iλ = y y ( i =, K, s) λ 0( ), s 0( i), i (3) whereis the itesity vectr, ad s represets -radial iput slacks vectr. Let a ptimal sluti f [SBM-I-C] be ( λ, s ). The, the bective fucti ca be rewritte as m i s i i τ =. m (4) Hece the SBM scre τ is the average f the cmpet-wise reducti rates which may vary frm e iput t ather. The SBM mdel is -radial. O the ther had, as ted i (2), the CCR scre θ satisfies the relatiship θ = Xλ + s = + s. Hece, we have i + si i θ = ( i). (5) The cmpet-wise reducti rates are the same fr all iputs. This same prprtial reducti rate, i.e. radial reducti rate, is the CCR scre. Betwee the SBM τ ad the CCR θ we have the iequality τ θ. See Te [8] fr mre details f their cmpariss. 3

4 GRIPS Plicy Ifrmati Ceter Discussi Paper : Shrtcmigs f the radial ad -radial mdels I this secti we pit ut shrtcmigs f the radial ad -radial DEA mdels. (a) Shrtcmigs f the CCR Mdel The mai shrtcmig f the CCR mdel is the eglect f -radial slacks s i reprtig f the efficiecy scre θ. I may cases, we fid a lt f remaiig -radial slacks. S, if these slacks have a imprtat rle i evaluatig maagerial efficiecy, the radial appraches may mislead the decisi whe we utilize the efficiecy scre θ as the ly ide fr evaluatig perfrmace f DMUs. Furthermre, as t the prprtial chage θ, if we emply labr, materials ad capital as iputs, sme f them are substitutial ad d t chage prprtially. The radial (CCR) mdel cat cpe with such cases prperly. (b) Shrtcmigs f the SBM Mdel Sice mdels such as SBM capture the -radial slacks directly, the ptimal efficiecy value τ accuts fr the -radial slacks which are t csidered i the radial mdels. The SBM-precti t the efficiet frtier is defied by = s. Thus, the prected DMU may lse the prprtiality i the rigial because s is t ecessarily prprtial t. This is characteristic f the -radial mdels, ad if the lss f the rigial prprtiality is iapprpriate fr the aalysis, the this becmes a shrtcmig fr -radial mdels. Yet ather equally sigificat shrtcmig f SBM arises frm the ature f the liear prgrammig sluti, where the ptimal slacks ted t ehibit a sharp ctrast i takig psitive ad zer values. See Avkira et al. [4] fr mre detailed cmpariss f the zer ad -zer patters i the ptimal slacks i the SBM mdel. 3. A epsil-based measure f efficiecy (EBM) As pited ut i the precedig secti, bth radial ad -radial mdels have merits ad demerits regardig the prprtiality f the iputs/utputs chage. I this secti, we prpse a cmprmised mdel called epsil-based measure (EBM) which has bth radial ad -radial features i a uified framewrk. We defie the primal ad dual pair [EBM-I-C] ad [Dual] as fllws: [EBM-I-C] γ = mi θ ε θ, λ, s subect t θ m i i ws i (6) Xλ s = 0 (7) Yλ y λ 0 s 0. (8) [Dual] 4

5 GRIPS Plicy Ifrmati Ceter Discussi Paper : 09-2 γ = ma uy v,u (9) where subect t v (0) = vx + uy 0 () v i ε w i ( i =, K, m) (2) i u 0, wi is the weight (relative imprtace) f iput i ad satisfies m i w = ( w 0 i), ad i ε is a key parameter which cmbies the radial θ ad the -radial slacks terms. Parameters ε ad ws the term i i i w must be supplied prir t the efficiecy measuremets. As ca be see frm i the bective fucti f [EBM-I-C], i i s is uits-ivariat ad s wi shuld be a uits-ivariat value reflectig the relative imprtace f resurce i. We will discuss this subect i the succeedig sectis. [Prpsiti ] γ i [EBM-I-C] satisfies γ 0ad is uits-ivariat, i.e. γ is idepedet f the uits i which the iputs ad utputs are measured. [Prpsiti 2] If we set ε = 0 i [EBM-I-C], the it reduces t the iput-rieted CCR mdel. [Prpsiti 3] If we set θ = ad ε = i [EBM-I-C], the it reduces t the iput-rieted SBM mdel. Thus, [EBM-I-C] icludes the radial CCR ad the -radial SBM mdels as special cases, but it is basically -radial. The cstraits (0) ad (2) lead t m v i i ε = v =. Thus, ε must be t greater tha uity. [Prpsiti 4] [EBM-I-C] ad [Dual] have a fiite ptima fr ε [0,]. [Prpsiti 5] Fr ε >, [Dual] has feasible sluti ad [EBM-I-C] has ubuded sluti. [Prpsiti 6] γ is -icreasig i ε. [Defiiti ] (EBM iput-efficiecy) DMU is called EBM iput-efficiet if γ =. 5

6 GRIPS Plicy Ifrmati Ceter Discussi Paper : 09-2 [Defiiti 2] (EBM precti) Let a ptimal sluti t (6)-(8) be ( θ, λ, s ). We defie the precti f DMU (, y ) as fllws. [Prpsiti 7] = = θ = Yλ. Xλ s y The prected DMU (, y ) is EBM iput-efficiet. (See Appedi A fr a prf.) (3) [EBM-I-C] ca be maipulated i ather frm by itrducig a variable γ = mi ( ε ) θ + ε θ,, λ, s subect t Xλ = 0 = θ s Yλ y λ 0s, 0. m wi i i = θ s as fllws. This frmulati idicates that γ is btaied as the ptimal iterally dividig value betwee the radial θ ad the -radial term m wi i / i. Sice θ is t restricted, its ptimal value ca be greater tha, ad hece the ptimal is t ecessarily less tha r equal t. We tice that the cmpsite sigle stage apprach like the EBM was cmmeted i Ali ad Seifrd [3] ad further develped by Jhs ad Ruggier [7]. Hwever, ur bective is quite differet frm the precedig es as ca be see i the fllwig sectis. 4. Hw t determie epsil ad weights I EBM, the values f ε ad (4) w play the cetral rle fr evaluatig efficiecy f DMUs. We wuld like t determie them frm the data set (X, Y), sice DEA is a data drive methd. I this secti firstly we bserve tw etreme cases. The we itrduce a affiity ide betwee tw vectrs which replaces the Pears s crrelati cefficiet. 4. Tw etreme cases () Narrw rage case Figure plts a eample f data ccerig iputs ad 2 ccetratig i a arrw rage. If all iputs ad utputs g alg with the similar behavir, the assumpti f prprtial (radial) mdel ca be effected. Thus, i such case, we have ε 0ad the CCR mdel is a valid chice. θ 6

7 GRIPS Plicy Ifrmati Ceter Discussi Paper : Figure : Narrw rage case (2) Widely scattered case I the ther etreme case, if the bserved data scatters widely as eemplified i Figure 2, the -radial mdel ca be applied. Thus, we have ε ad the SBM mdel with θ = is a chice althugh we d t shut ff the assumpti f radial mdels depedig the characteristics f prblems. 2 0 Figure 2: Widely scattered case These etreme cases suggest that ε ca be determied i the ctet f the degree f crrelatis amg iputs (utputs). Several authrs, e.g. Ueda ad Hshiai [9] ad Adler ad Glay [, 2] amg thers, utilized crrelati matri f iputs (utputs) ad applied pricipal cmpet aalysis (PCA) t DEA. Their mai bectives were itegrati f iputs (utputs) t ther represetative idicatrs. 7

8 GRIPS Plicy Ifrmati Ceter Discussi Paper : 09-2 I this paper, we emply similar but differet crrelati matri as described i the et secti i rder t gauge affiity amg iputs which will be utilized t estimate parameters ε ad i the EBM. 4.2 Diversity ide ad affiity ide Let R ad R a b be tw psitive vectrs with dimesi. They represet bserved values fr + + certai iput items ver DMUs. We defie a affiity ide S(, ab) betwee a ad b with the fllwig prperties. (P) S ( aa, ) = ( a ) Idetical (P2) S( ab, ) = S( ba, ) Symmetric (P3) St ( ab, ) = S( ab, )( t> 0) Uits-ivariat ad (P4) S( ab, ) 0 ( ab, ). The usual Pears s crrelati cefficiet itrduces the traslati f rigi i calculatig crrelatis. I ur mdel, we wish t evaluate affiity f tw vectrs withut traslati f rigi. Therefre, we itrduce ather crrelati cefficiet called affiity ide. Let us defie w b c = l ( =, K, ) a c= ma c = { } mi { } c = ma c ad c = mi c. (5) [Defiiti 3] (diversity ide) We defie the diversity ide f vectrs a ad b as the deviati f { c } frm the average c i the fllwig way. c c = D( a,b ) = (6) c ( c ) ma ma mi = 0 if c = c. mi [Prpsiti 8] 2 0 D( a,b) = D( b,a ). (7) See Appedi B fr a prf. D ( a,b) = 0ccurs if ad ly if tw vectrs a ad b are prprtial. [Defiiti 4] (affiity ide) We defie the affiity ide S( a,b) betwee tw vectrs a ad b by S( a,b) = 2 D( a,b ). (8) [Prpsiti 9] It hlds S( a,b ) 0. S( a,b ) satisfies prperties (P), (P2), (P3) ad (P4). 8

9 GRIPS Plicy Ifrmati Ceter Discussi Paper : 09-2 The reas why we emply the affiity ide (5) istead f the Pears s crrelati cefficiet is the fllwig:. Pears s crrelati cefficiet is defied by ( a a)( b b) = r( a,b ) =, 2 2 ( a a) ( b b) = = where aad bare respectively averages f { a} ad { b }. I this frmula, the abslute magitude f a ad b b effects r( a,b) strgly. I ctrast, i DEA, the relative measure, e.g. a, is a mai ccer. 2. Pears s crrelati cefficiet results i the rage r( a,b ). Hece, i the pricipal cmpet aalysis we will utilize i the et secti, it is t guarateed that the pricipal vectr csists f -egative cmpets. Althugh it is pssible t adust r( a,b) it [0, ], this might brig a skew distributi, sice mst f r( a,b ) are -egative i DEA applicatis. 3. We emply the lgarithmic fucti l b / a istead f b / a, because the latter vilates the prperty (P2). 5. Use f affiity matri i EBM I this secti, we measure the diversity f prducti pssibility set by meas f the affiity matri derived frm the bserved iputs ad utputs. Althugh we describe the methd i the iput-rieted mdel uder the cstat returs-t-scale (CRS) assumpti, we ca mdify it t the utput-rieted ad -rieted mdels uder cstat r variable returs-t-scale (VRS) assumptis. We discuss this subect i Secti 8. Step. Creati f prected VRS-efficiet DMUs I mst DEA mdels, the prducti pssibility set is spaed by the efficiet DMUs which usually csist f a small prti f the etire DMUs. I rder t icrease the accuracy f ur estimati, we first prect all DMUs t the VRS (variable returs-t-scale)-efficiet frtiers usig the Additive mdel r -rieted SBM mdel belw. We ca emply the bserved data (X, Y) istead f the prected DMUs i this step. Hwever, we utilized the prected DMUs, because ur mai ccers are the shape f frtiers. 9

10 GRIPS Plicy Ifrmati Ceter Discussi Paper : 09-2 [ADD] ma subect t m s + si s + i i yi i iλ i = = + s ( i =, K, m) + i iλ i = y = y s ( i =, K, s) = λ = + i i λ 0 ( ), s 0 ( i), s 0( i). (9) [SBM] m mi + s subect t m s i i s + si yi i iλ i = = + s ( i =, K, m) + i iλ i = y = y s ( i =, K, s) = λ = + i i λ 0( ), s 0( i), s 0( i). (20) Usig the ptimal slacks s ad s + we defie the prected iput ad utput fr DMU by i i i + i i i = s ( i =, K, m) y = y + s ( i =, K, s). We tice that [ADD] ad [SBM] may prduce differet prectis but they are the efficiet frtiers f the prducti pssibility set. Thus, we have VRS-efficiet DMUs deted by (2) L LLLL L X ml m m = = y y Y L y LLLL L ysl y s ys (22) All CRS (cstat-returs-t-scale) efficiet DMUs are icluded i this set alg with VRS-efficiet DMUs. Step 2. Frmati f affiity matri I the iput-rieted case, we calculate the affiity matri m m S = si R with the elemets 0

11 GRIPS Plicy Ifrmati Ceter Discussi Paper : 09-2 All elemets f the matri S satisfy the buds: s = S(, ) ( i, =, K, m) (23) i i s 0( ( i)). (24) i Step 3. Slvig the largest eigevalue ad eigevectr f the affiity matri By the frmati rule, S is symmetric ad -egative with the diagal elemets equal t uity. It has m pairs f eigevalue ad eigevectr. By the Perr-Frbeius therem fr -egative matrices, S has the largest eigevalue ρ with its assciated -egative eigevectr w ( 0 ). The -egative w crrespds t the weight f iput factrs. Sice S is -egative defiite, we have m ρ. Step 4. Calculati f ε ad w fr the EBM We defie ε ad w i the EBM as fllws. m ρ ε = (if m > ) m = 0 (if m = ). (25) w =. m wi w (26) The thus defied ε ad w satisfy the relatiship 0 ε ad ew =. Step 5. Use f ε ad w i the EBM These parameters are utilized i the EBM mdel [EBM-I-C]. 6. Ratiale f the prpsed EBM I this secti we demstrate the ratiale f the scheme prpsed i the precedig secti. Befre gig it theretical discussis, we shw sme real wrld data ccerig iput/utput items. Figure 4 depicts 84 samples f. f dctrs (as iput) vs.. f beds (as iput) f Japaese muicipally-wed hspitals. Figure 5 shws. f dctrs (as iput) vs. reveue/day (as utput) i the same 84 sample hspitals. Sice the muicipal hspitals are, t sme etet, stadardized uder the ctrl f respective admiistrative ffices, may iputs ad utputs have psitive relatiship ad hece the affiity matri is epected t have high affiity values. Csequetly, its pricipal eigevalue will be large ad hece ε will be small.

12 GRIPS Plicy Ifrmati Ceter Discussi Paper : 09-2 Figure 4: Empirical data Figure 5: Empirical data 2 Figures 6 ad 7 ehibit tw pltted data ccerig. f emplyees vs.. f visitrs ad area vs.. f visitrs fr Japaese regial museums. Sice museum busiess is t stadardized cmpared with regial hspitals, they are distributed widely. I this case, the affiity matri is epected t csist f lw values with relatively small pricipal eigevalue ad hece large ε. 2

13 GRIPS Plicy Ifrmati Ceter Discussi Paper : 09-2 Figure 6: Empirical data 3 Figure 7: Empirical data 4 Figures 8 ad 9 plt data f 273 electric pwer plats i the U.S. ccerig the geeratig pwer capacity (GW) (iput) vs.. f emplyees (iput) ad the csumed fuel (milli BTU) (iput) vs.. f emplyees (iput). They are psitively crrelated but csiderably diversified. 3

14 GRIPS Plicy Ifrmati Ceter Discussi Paper : 09-2 N. f emplyees GW Figure 8: Empirical data 5 N. f emplyees milli BTU Figure 9: Empirical data 6 T The ellipsid w Sw = has the pricipal ais i the first (psitive) quadrat as eemplified i Figure 0. As the degree f affiity becmes higher ad higher, the shape f the ellipsid cmes t be flat ad the largest eigevalue ρ teds t m. Thus, ε i (25) teds t 0. This cmes clse t the CCR mdel, i.e. all iputs ad utputs fllw prprtial chages. 4

15 GRIPS Plicy Ifrmati Ceter Discussi Paper : Figure 0: Ellipsid f affiity matri Depedig the degree f affiity amg iputs, the pricipal eigevalue Cversely, the mre the data scatters widely, the mre ρ teds t ad the mre ρ icreases up t m. ε grws up. Hece, the mdel behaves SBM-like. Therefre, it ca safely be said that ε cdeses the affiity matri i a sigle value reflectig the scatterig f the data set. We w tur t the psitive eigevectr w crrespdig t the eigevalue ρ. First f all, we tice that w is uits-ivariat, sice the affiity matri is uits-ivariat. Suppse that, i the affiity matri = ( s i ) S, s > s2 0( = 3, K, m), the it hlds that w > w2. This idicates that the item which has higher affiity with thers has a large prti i the eigevectr, whereas item i with urelated t thers, i.e., s i 0( i) has wi 0. Thus, the magitude f elemets f w idicates imprtace f the item amg the whle items. We ca stregthe the discrimiati pwer efficiecy by impsig weight t slacks i prprti t applicati f the pricipal cmpet aalysis (PCA) t DEA. We te here that, i the iput-rieted mdel, we estimate w. Thus, this scheme is a ε depedig ly the iput data X, but t the utput data Y. This meas that the bective fucti i [EBM-I-C] relates t the radial factrθ ad the diversity idicatr ε. The frmer represets the radial feature f iputs ad the later implies the -radial characteristics f iputs. The iteractis betwee iput X ad utput Y are described i the cstraits f [EBM-I-C] thrugh the itermediary f the itesity vectr 7. Illustrative eamples I this secti, we eplai the EBM usig three eamples ad cmpare the results with the radial 5

16 GRIPS Plicy Ifrmati Ceter Discussi Paper : 09-2 (CCR) ad -radial (SBM) scres. 7. Eample Table reprts cmpariss f CCR-I, SBM-I-C ad EBM-I-C scres fr si DMUs A, B, C, D, E ad F with tw iputs (, 2 ) ad a sigle utput (y = ). Figure plts them graphically. This figure idicates that the data are ccetrated i a arrw gauge. See als Figure 2. Table : 2 y CCR-I SBM-I-C EBM-I-C A B C D E F Figure : Eample Figure 2: Cmparis f scres As ca be see, EBM scres are the same with the CCR scres. We illustrate the EBM scheme i rder. Step : We used [ADD] fr fidig slacks ad prected DMUs t efficiet frtiers, as shw i 6

17 GRIPS Plicy Ifrmati Ceter Discussi Paper : 09-2 Table 2. They are all prected t the ly e efficiet DMU A. Table 2: Prected DMUs 2 y A B C D E F Step 2: We calculated the diversity matri by the frmula (6). See Table 3. Sice the set f efficiet DMUs csists f ly e DMU A, diversity eists. Table 3: Diversity matri fr Eample Step 3: The affiity matri is calculated by the frmula (23) ad displayed i Table 4. Table 4: Affiity matri fr Eample 2 2 Step 4: The largest eigevalue ad eigevectr f the affiity matri are ρ = 2 ad w = (0.5, 0.5). Hece we have ε I this case, the ellipsid is perfectly flat. = ( mρ) / ( m ) = 0, w = 0.5, w2 = 0.5. Step 5: Usig these parameter values we applied EBM-I-C t the si DMUs. Sice we have ε = 0, the scres are idetical with the CCR scres. 7.2 Eample 2 This eample has diversified DMUs as ehibited i Table 5 ad Figure 3. Table 5: Eample 2 2 y CCR-I SBM-I-C EBM-I-C 7

18 GRIPS Plicy Ifrmati Ceter Discussi Paper : 09-2 A 2 6 B 6 3 C D Figure 3: Eample 2 The prected data are ehibited i Table 6. Table 6: Prected data fr Eample 2 2 y A 2 6 B 6 3 C 6 3 D 2 6 The diversity matri fr the EBM-I-C mdel is displayed i Table 7 alg with the affiity matri i Table 8. Table 7: Diversity matri fr Eample

19 GRIPS Plicy Ifrmati Ceter Discussi Paper : 09-2 Table 8: Affiity matri fr Eample The largest eigevalue ad eigevectr f the affiity matri are ρ = ad w = (0.5,0.5). Hece we have ε = ( mρ )/( m ) = w = 0.5, = 0.5. w2 The value ε = is the largest e shwig the diversity f the data set ad EBM-I-C results are idetical with the SBM results. 7.3 Eample 3 Table 9 reprts efficiecy scres f 2 hspitals. We utilized umbers f dctrs ad urses as iputs, ad umbers f utpatiets ad ipatiets per mth as utputs. Figure 4 displays cmpariss f three scres: CCR-I, SBM-I-C ad EBM-I-C. Table 9: Hspital data ad efficiecy scres (I)Dctr (I)Nurse (O)Outpatiet (O)Ipatiet CCR-I SBM-I-C EBM-I-C A B C D E F G H I J K L

20 GRIPS Plicy Ifrmati Ceter Discussi Paper : 09-2 Figure 4: Cmpariss f scres (hspital) We utilized [ADD] fr prectig the dataset t the VRS-efficiet frtiers ad btaied the ew dataset ehibited i Table 0. Table 0: Prected DMUs (hspital) Dctr Nurse Outpatiet Ipatiet A B C D E F G H I J K L The diversity matri is displayed i Table alg with the affiity matri i Table 2. Table : Diversity matri fr Eample 3 Dctr Nurse 20

21 GRIPS Plicy Ifrmati Ceter Discussi Paper : 09-2 Dctr Nurse Table 2: Affiity matri fr Eample 3 Dctr Nurse Dctr 0.47 Nurse 0.47 This affiity matri has the largest eigevalue ad eigevectr: ρ =.47, w = (0.5,0.5) Hece we have: ε = ( mρ) / ( m ) = w = 0.5, w2 = 0.5. The EBM scres were btaied usig these ε ad w values. Table 3 ehibits θ ad slacks s, s 2, s + ad s + i the sluti f the EBM-I-C mdel. It is 2 iterestig t tice that hspital D is iefficiet with the scre 0.986, i ctrast t the CCR ad SBM scre (efficiet). The EBM mdel impses restricti θ, ad D has a ptimalθ =.06( > ). Thus the ptimal sluti isists that all iputs are multiplied by.06 ad further. f dctr is decreased by the slacks s = The prected iputs fr D are = fr Dctr ad = fr Nurse. D s refereces are A ( λ A = 0.28 ) ad B ( λ B =.0588 ). D is recmmeded t reduce dctrs frm 27 t 24 ad icrease urses frm 68 t 7 i rder t imprve efficiecy. This is e f characteristics f the cmpsite mdel EBM, whereas such substituti f iputs cat ccur i the CCR r the SBM mdels. Table 3: θ ad slacks DMU Scre Rak θ s s2 s+ s 2 + A B C D E F G

22 GRIPS Plicy Ifrmati Ceter Discussi Paper : 09-2 H I J K L Etesis t ther rietatis ad variable returs-t-scale mdels S far, we have develped the EBM i the iput-rietati uder the cstat returs-t-scale evirmet. Hwever, we ca eted it t ther rietatis ad returs-t-scale evirmet as fllws. I every variati we fllw Step i Secti 5 ad btai the set f prected VRS-efficiet DMUs (22). 8. Output-rieted EBM Step 2. Frmati f affiity matri I the utput-rieted case, we calculate the affiity matri s s S = si R with the elemets si = S( yi, y ) ( i, =, K, s) (27) Step 3. Slvig the largest eigevalue ad eigevectr f the affiity matri We slve the largest eigevalue ρ y ad eigevectr w y f the affiity matri S i (27). + Step 4. Calculati f ε y ad w. We defie s ρ y ε y = (if s > ), = 0 (if s = ) s + w y w =. s wyi (28) Usig ε y [EBM-O-C] + ad w we slve the fllwig liear prgram: subect t = ma / τ η εy η, λ, s+ + s + + i i w s y i (29) Xλ (30) y Yλ + s λ 0 + s 0. η + = 0 (3) 8.2 N-rieted (bth-rieted) EBM We apply Steps 2 ad 3 fr the iput-rieted ad the utput-rieted affiity matri separately, ad 22

23 GRIPS Plicy Ifrmati Ceter Discussi Paper : btai ε, ε, w, ad w. The -rieted EBM ca be frmulated i the fllwig fractial y prgram which ca be slved as a liear prgram usig the Chares-Cper trasfrmati. (See Cper et al. [6].) κ = mi + θη,, λ,s, s subect t θ Xλ s = 0 θ ε η + ε m ηy Yλ + s = 0 λ 0s, 0s, 0. y Variable returs-t-scale EBM All mdels ca be mdified t variable returs-t-scale (VRS) es by addig the cditi: λ+ λ2 + L + λ =. (33) 9. Ccludig remarks I this paper, we have prpsed EBM as a third ple f techical efficiecy i DEA by cmbiig radial ad -radial mdels i a uified framewrk. Sice DEA is a data drive methd, we eed t measure techical efficiecy frm the bserved data uder less assumptis its distributi. Fr this purpse we itrduced a ew ide called affiity ide fr measurig similarity betwee tw vectrs fr use i DEA. Usig this ide, we defied a scalar measure epsil ( ε ) that represets the diversity r the scatterig f the bserved dataset. We prpsed a scheme fr settig weights t slacks based the pricipal cmpet aalysis. We als eteded it t ther rietatis ad returs-t-scale assumptis. Future research subects iclude search fr ther measure f affiity ide that satisfies the prperties (P) t (P4), etesis t Super-EBM ad idetificatis f returs-t-scale ad scale efficiecy uder this mdel. Refereces [] Adler N, Glay B. (200) Evaluati f deregulated airlie etwrks usig data evelpmet aalysis cmbied with pricipal cmpet aalysis with a applicati t Wester Eurpe. Eurpea Jural f Operatial Research, 32, [2] Adler N, Glay B. (2002) Icludig pricipal cmpet weights t imprve discrimiati i data evelpmet aalysis. Jural f the Operatial Research Sciety, 53, [3] Ali AL, Seifrd LM. (993) Cmputatial accuracy ad ifiitesimals i data evelpmet aalysis. INFOR, 3, [4] Avkira N, Te K, Tsutsui M. (2008) Bridgig radial ad -radial measures f efficiecy i DEA. Aals f Operatis Research, 64, [5] Chares A, Cper WW, Rhdes E. (978) Measurig the efficiecy f decisi makig uits. Eurpea Jural f Operatial Research, 2, s i i ws i + + i i w s y i (32) 23

24 GRIPS Plicy Ifrmati Ceter Discussi Paper : 09-2 [6] Cper WW, Seifrd LM, Te K. (2007) Data Evelpmet Aalysis: A Cmprehesive Tet with Mdels, Applicatis, Refereces ad DEA-Slver Sftware, Secd Editi, Spriger. [7] Jhs AL ad Ruggier J. (2009) - Substitutability, slacks ad data evelpmet aalysis. Wrkig paper. [8] Te K. (200) A slacks-based measure f efficiecy i data evelpmet aalysis. Eurpea Jural f Operatial Research, 30, [9] Ueda T, Hshiai Y. (997) Applicati f pricipal cmpet aalysis fr parsimius summarizati f DEA iputs ad/r utputs. Jural f Operatis Research Sciety f Japa, 40, Appedi A. Prf f Prpsiti 7 Sice (, y ) is EBM iput-iefficiet, it hlds that Let a ptimal sluti fr (, y ) be The crrespdig cstraits fr (, y ) are: This reduces t: θ i i m ws γ = θ ε <. (A) i ( γ, θ, λ, s ). The EBM bective fucti value is: i i i m ws γ = θ ε. (A2) = Xλ + s, y Yλ. (A3) = Xλ + s + θ s, y y Yλ. (A4) θ θ This is ather epressi fr (, y ) ad its bective fucti value is: m w ( ) m i si + θ si wi si i i f = θ θ ε = θ γ ε. (A5) We have three pssibilities as fllws: i) The case θ <. I this case, it hlds that f < γ. This ctradicts the ptimality f γ fr (, y ). Thus, this case ever ccurs. ii) The case θ =. I this case, by the ptimality f γ fr (, y ), we have s = 0( i). Thus, γ = ad (, y ) is EBM iput-efficiet. iii) The case θ >. Frm the ptimality f γ fr (, y ), it hlds that Hece we have m ws i i i θ γ ε γ m ws i i i ε θ +. (A6) γ i 24

25 GRIPS Plicy Ifrmati Ceter Discussi Paper : 09-2 Suppse that (, y ) is EBM-iefficiet, i.e. have: i i i m ws γ = θ ε <. The we We cmpare the terms i i i m ws θ < + ε. (A7) γ i i (A6) ad i i (A7). Sice i = θ i si, we have m ws ( ) k k γ i i = γ θ i si =εi si 0. k = k Thus, it hlds that γ i i. (A8) Cmparig (A6) ad (A7), we have m m ws i i ws i i + + > i γ = i i (, ) (A9) θ ε ε θ. This cat ccur. Thus, i this case, y is EBM iput-efficiet. Q.E.D. Appedi B. Prf f Prpsiti 8 (a) Prf f D(a,b)=D(b,a) Let d l( a / b ) (,, )) = = K, dma ma { d}, dmi mi{ d} ma mi mi ma = =, ad d = d. The it hlds that d = c ( =, K, ), d = c, d = c,ad d =c. Hece, we have d d c + c D( b,a) = = = D( a,b ). (B) d ( ma dmi ) ( cmi + cma ) (b) Prf f D(a,b) /2 If a ad b are prprtial, the it hlds that c ma = c mi ad D( a,b ) = 0. Otherwise if a ad b are t prprtial, the c ma > c mi ad D( a,b ) > 0. Let N ad N 2 be respectively the set f such that c cad c > c, ad = N ad 2 = N 2. We have = + 2. The umeratr f D(a,b) ca be trasfrmed it the fllwig: c c = ( c + c) + ( c c) = N N2 ( cmi + c) + 2( cma c) 2 2 = ( cma cmi )( ). 2 The last term i the last epressi attais the maimum /4at = /2 Hece, we have (B2) c c = Dab (, ) =. c ( c ) 2 ma mi (B3) D( a,b) /2 hlds whe { c } distributes as eemplified i Figure B. 25

26 GRIPS Plicy Ifrmati Ceter Discussi Paper : 09-2 c ma c c mi Figure B: The case D( a,b) /2 26

Super-efficiency Models, Part II

Super-efficiency Models, Part II Super-efficiec Mdels, Part II Emilia Niskae The 4th f Nvember S steemiaalsi Ctets. Etesis t Variable Returs-t-Scale (0.4) S steemiaalsi Radial Super-efficiec Case Prblems with Radial Super-efficiec Case

More information

Multi-objective Programming Approach for. Fuzzy Linear Programming Problems

Multi-objective Programming Approach for. Fuzzy Linear Programming Problems Applied Mathematical Scieces Vl. 7 03. 37 8-87 HIKARI Ltd www.m-hikari.cm Multi-bective Prgrammig Apprach fr Fuzzy Liear Prgrammig Prblems P. Padia Departmet f Mathematics Schl f Advaced Scieces VIT Uiversity

More information

Portfolio Performance Evaluation in a Modified Mean-Variance-Skewness Framework with Negative Data

Portfolio Performance Evaluation in a Modified Mean-Variance-Skewness Framework with Negative Data Available lie at http://idea.srbiau.ac.ir It. J. Data Evelpmet Aalysis (ISSN 345-458X) Vl., N.3, Year 04 Article ID IJDEA-003,3 pages Research Article Iteratial Jural f Data Evelpmet Aalysis Sciece ad

More information

D.S.G. POLLOCK: TOPICS IN TIME-SERIES ANALYSIS STATISTICAL FOURIER ANALYSIS

D.S.G. POLLOCK: TOPICS IN TIME-SERIES ANALYSIS STATISTICAL FOURIER ANALYSIS STATISTICAL FOURIER ANALYSIS The Furier Represetati f a Sequece Accrdig t the basic result f Furier aalysis, it is always pssible t apprximate a arbitrary aalytic fucti defied ver a fiite iterval f the

More information

Variations on the theme of slacks-based measure of efficiency in DEA

Variations on the theme of slacks-based measure of efficiency in DEA GRIPS Plicy Ifrmati Ceter Dicui Paper : 8-4 Variati the theme f lac-baed meaure f efficiecy i DEA Karu Te Natial Graduate Ititute fr Plicy Studie 7-22- Rppgi, Miat-u, Ty 6-8677, Japa te@gripacp Abtract:

More information

ENGI 4421 Central Limit Theorem Page Central Limit Theorem [Navidi, section 4.11; Devore sections ]

ENGI 4421 Central Limit Theorem Page Central Limit Theorem [Navidi, section 4.11; Devore sections ] ENGI 441 Cetral Limit Therem Page 11-01 Cetral Limit Therem [Navidi, secti 4.11; Devre sectis 5.3-5.4] If X i is t rmally distributed, but E X i, V X i ad is large (apprximately 30 r mre), the, t a gd

More information

Chapter 3.1: Polynomial Functions

Chapter 3.1: Polynomial Functions Ntes 3.1: Ply Fucs Chapter 3.1: Plymial Fuctis I Algebra I ad Algebra II, yu ecutered sme very famus plymial fuctis. I this secti, yu will meet may ther members f the plymial family, what sets them apart

More information

The Excel FFT Function v1.1 P. T. Debevec February 12, The discrete Fourier transform may be used to identify periodic structures in time ht.

The Excel FFT Function v1.1 P. T. Debevec February 12, The discrete Fourier transform may be used to identify periodic structures in time ht. The Excel FFT Fucti v P T Debevec February 2, 26 The discrete Furier trasfrm may be used t idetify peridic structures i time ht series data Suppse that a physical prcess is represeted by the fucti f time,

More information

Fourier Series & Fourier Transforms

Fourier Series & Fourier Transforms Experimet 1 Furier Series & Furier Trasfrms MATLAB Simulati Objectives Furier aalysis plays a imprtat rle i cmmuicati thery. The mai bjectives f this experimet are: 1) T gai a gd uderstadig ad practice

More information

A New Method for Finding an Optimal Solution. of Fully Interval Integer Transportation Problems

A New Method for Finding an Optimal Solution. of Fully Interval Integer Transportation Problems Applied Matheatical Scieces, Vl. 4, 200,. 37, 89-830 A New Methd fr Fidig a Optial Sluti f Fully Iterval Iteger Trasprtati Prbles P. Padia ad G. Nataraja Departet f Matheatics, Schl f Advaced Scieces,

More information

BIO752: Advanced Methods in Biostatistics, II TERM 2, 2010 T. A. Louis. BIO 752: MIDTERM EXAMINATION: ANSWERS 30 November 2010

BIO752: Advanced Methods in Biostatistics, II TERM 2, 2010 T. A. Louis. BIO 752: MIDTERM EXAMINATION: ANSWERS 30 November 2010 BIO752: Advaced Methds i Bistatistics, II TERM 2, 2010 T. A. Luis BIO 752: MIDTERM EXAMINATION: ANSWERS 30 Nvember 2010 Questi #1 (15 pits): Let X ad Y be radm variables with a jit distributi ad assume

More information

Quantum Mechanics for Scientists and Engineers. David Miller

Quantum Mechanics for Scientists and Engineers. David Miller Quatum Mechaics fr Scietists ad Egieers David Miller Time-depedet perturbati thery Time-depedet perturbati thery Time-depedet perturbati basics Time-depedet perturbati thery Fr time-depedet prblems csider

More information

Study of Energy Eigenvalues of Three Dimensional. Quantum Wires with Variable Cross Section

Study of Energy Eigenvalues of Three Dimensional. Quantum Wires with Variable Cross Section Adv. Studies Ther. Phys. Vl. 3 009. 5 3-0 Study f Eergy Eigevalues f Three Dimesial Quatum Wires with Variale Crss Secti M.. Sltai Erde Msa Departmet f physics Islamic Aad Uiversity Share-ey rach Ira alrevahidi@yah.cm

More information

Solutions. Definitions pertaining to solutions

Solutions. Definitions pertaining to solutions Slutis Defiitis pertaiig t slutis Slute is the substace that is disslved. It is usually preset i the smaller amut. Slvet is the substace that des the disslvig. It is usually preset i the larger amut. Slubility

More information

Axial Temperature Distribution in W-Tailored Optical Fibers

Axial Temperature Distribution in W-Tailored Optical Fibers Axial Temperature Distributi i W-Tailred Optical ibers Mhamed I. Shehata (m.ismail34@yah.cm), Mustafa H. Aly(drmsaly@gmail.cm) OSA Member, ad M. B. Saleh (Basheer@aast.edu) Arab Academy fr Sciece, Techlgy

More information

Fourier Method for Solving Transportation. Problems with Mixed Constraints

Fourier Method for Solving Transportation. Problems with Mixed Constraints It. J. Ctemp. Math. Scieces, Vl. 5, 200,. 28, 385-395 Furier Methd fr Slvig Trasprtati Prblems with Mixed Cstraits P. Padia ad G. Nataraja Departmet f Mathematics, Schl f Advaced Scieces V I T Uiversity,

More information

[1 & α(t & T 1. ' ρ 1

[1 & α(t & T 1. ' ρ 1 NAME 89.304 - IGNEOUS & METAMORPHIC PETROLOGY DENSITY & VISCOSITY OF MAGMAS I. Desity The desity (mass/vlume) f a magma is a imprtat parameter which plays a rle i a umber f aspects f magma behavir ad evluti.

More information

Intermediate Division Solutions

Intermediate Division Solutions Itermediate Divisi Slutis 1. Cmpute the largest 4-digit umber f the frm ABBA which is exactly divisible by 7. Sluti ABBA 1000A + 100B +10B+A 1001A + 110B 1001 is divisible by 7 (1001 7 143), s 1001A is

More information

ENGI 4421 Central Limit Theorem Page Central Limit Theorem [Navidi, section 4.11; Devore sections ]

ENGI 4421 Central Limit Theorem Page Central Limit Theorem [Navidi, section 4.11; Devore sections ] ENGI 441 Cetral Limit Therem Page 11-01 Cetral Limit Therem [Navidi, secti 4.11; Devre sectis 5.3-5.4] If X i is t rmally distributed, but E X i, V X i ad is large (apprximately 30 r mre), the, t a gd

More information

Physical Chemistry Laboratory I CHEM 445 Experiment 2 Partial Molar Volume (Revised, 01/13/03)

Physical Chemistry Laboratory I CHEM 445 Experiment 2 Partial Molar Volume (Revised, 01/13/03) Physical Chemistry Labratry I CHEM 445 Experimet Partial Mlar lume (Revised, 0/3/03) lume is, t a gd apprximati, a additive prperty. Certaily this apprximati is used i preparig slutis whse ccetratis are

More information

, the random variable. and a sample size over the y-values 0:1:10.

, the random variable. and a sample size over the y-values 0:1:10. Lecture 3 (4//9) 000 HW PROBLEM 3(5pts) The estimatr i (c) f PROBLEM, p 000, where { } ~ iid bimial(,, is 000 e f the mst ppular statistics It is the estimatr f the ppulati prprti I PROBLEM we used simulatis

More information

Grade 3 Mathematics Course Syllabus Prince George s County Public Schools

Grade 3 Mathematics Course Syllabus Prince George s County Public Schools Ctet Grade 3 Mathematics Curse Syllabus Price Gerge s Cuty Public Schls Prerequisites: Ne Curse Descripti: I Grade 3, istructial time shuld fcus fur critical areas: (1) develpig uderstadig f multiplicati

More information

A Hartree-Fock Calculation of the Water Molecule

A Hartree-Fock Calculation of the Water Molecule Chemistry 460 Fall 2017 Dr. Jea M. Stadard Nvember 29, 2017 A Hartree-Fck Calculati f the Water Mlecule Itrducti A example Hartree-Fck calculati f the water mlecule will be preseted. I this case, the water

More information

A Study on Estimation of Lifetime Distribution with Covariates Under Misspecification

A Study on Estimation of Lifetime Distribution with Covariates Under Misspecification Prceedigs f the Wrld Cgress Egieerig ad Cmputer Sciece 2015 Vl II, Octber 21-23, 2015, Sa Fracisc, USA A Study Estimati f Lifetime Distributi with Cvariates Uder Misspecificati Masahir Ykyama, Member,

More information

K [f(t)] 2 [ (st) /2 K A GENERALIZED MEIJER TRANSFORMATION. Ku(z) ()x) t -)-I e. K(z) r( + ) () (t 2 I) -1/2 e -zt dt, G. L. N. RAO L.

K [f(t)] 2 [ (st) /2 K A GENERALIZED MEIJER TRANSFORMATION. Ku(z) ()x) t -)-I e. K(z) r( + ) () (t 2 I) -1/2 e -zt dt, G. L. N. RAO L. Iterat. J. Math. & Math. Scl. Vl. 8 N. 2 (1985) 359-365 359 A GENERALIZED MEIJER TRANSFORMATION G. L. N. RAO Departmet f Mathematics Jamshedpur C-perative Cllege f the Rachi Uiversity Jamshedpur, Idia

More information

Ch. 1 Introduction to Estimation 1/15

Ch. 1 Introduction to Estimation 1/15 Ch. Itrducti t stimati /5 ample stimati Prblem: DSB R S f M f s f f f ; f, φ m tcsπf t + φ t f lectrics dds ise wt usually white BPF & mp t s t + w t st. lg. f & φ X udi mp cs π f + φ t Oscillatr w/ f

More information

Mean residual life of coherent systems consisting of multiple types of dependent components

Mean residual life of coherent systems consisting of multiple types of dependent components Mea residual life f cheret systems csistig f multiple types f depedet cmpets Serka Eryilmaz, Frak P.A. Cle y ad Tahai Cle-Maturi z February 20, 208 Abstract Mea residual life is a useful dyamic characteristic

More information

5.1 Two-Step Conditional Density Estimator

5.1 Two-Step Conditional Density Estimator 5.1 Tw-Step Cditial Desity Estimatr We ca write y = g(x) + e where g(x) is the cditial mea fucti ad e is the regressi errr. Let f e (e j x) be the cditial desity f e give X = x: The the cditial desity

More information

Comparative analysis of bayesian control chart estimation and conventional multivariate control chart

Comparative analysis of bayesian control chart estimation and conventional multivariate control chart America Jural f Theretical ad Applied Statistics 3; ( : 7- ublished lie Jauary, 3 (http://www.sciecepublishiggrup.cm//atas di:.648/.atas.3. Cmparative aalysis f bayesia ctrl chart estimati ad cvetial multivariate

More information

Active redundancy allocation in systems. R. Romera; J. Valdés; R. Zequeira*

Active redundancy allocation in systems. R. Romera; J. Valdés; R. Zequeira* Wrkig Paper -6 (3) Statistics ad Ecmetrics Series March Departamet de Estadística y Ecmetría Uiversidad Carls III de Madrid Calle Madrid, 6 893 Getafe (Spai) Fax (34) 9 64-98-49 Active redudacy allcati

More information

ALE 26. Equilibria for Cell Reactions. What happens to the cell potential as the reaction proceeds over time?

ALE 26. Equilibria for Cell Reactions. What happens to the cell potential as the reaction proceeds over time? Name Chem 163 Secti: Team Number: AL 26. quilibria fr Cell Reactis (Referece: 21.4 Silberberg 5 th editi) What happes t the ptetial as the reacti prceeds ver time? The Mdel: Basis fr the Nerst quati Previusly,

More information

Directional Duality Theory

Directional Duality Theory Suther Illiis Uiversity Carbdale OpeSIUC Discussi Papers Departmet f Ecmics 2004 Directial Duality Thery Daiel Primt Suther Illiis Uiversity Carbdale Rlf Fare Oreg State Uiversity Fllw this ad additial

More information

are specified , are linearly independent Otherwise, they are linearly dependent, and one is expressed by a linear combination of the others

are specified , are linearly independent Otherwise, they are linearly dependent, and one is expressed by a linear combination of the others Chater 3. Higher Order Liear ODEs Kreyszig by YHLee;4; 3-3. Hmgeeus Liear ODEs The stadard frm f the th rder liear ODE ( ) ( ) = : hmgeeus if r( ) = y y y y r Hmgeeus Liear ODE: Suersiti Pricile, Geeral

More information

MODIFIED LEAKY DELAYED LMS ALGORITHM FOR IMPERFECT ESTIMATE SYSTEM DELAY

MODIFIED LEAKY DELAYED LMS ALGORITHM FOR IMPERFECT ESTIMATE SYSTEM DELAY 5th Eurpea Sigal Prcessig Cferece (EUSIPCO 7), Pza, Plad, September 3-7, 7, cpyright by EURASIP MOIFIE LEAKY ELAYE LMS ALGORIHM FOR IMPERFEC ESIMAE SYSEM ELAY Jua R. V. López, Orlad J. bias, ad Rui Seara

More information

Matching a Distribution by Matching Quantiles Estimation

Matching a Distribution by Matching Quantiles Estimation Jural f the America Statistical Assciati ISSN: 0162-1459 (Prit) 1537-274X (Olie) Jural hmepage: http://www.tadflie.cm/li/uasa20 Matchig a Distributi by Matchig Quatiles Estimati Niklas Sgurpuls, Qiwei

More information

AP Statistics Notes Unit Eight: Introduction to Inference

AP Statistics Notes Unit Eight: Introduction to Inference AP Statistics Ntes Uit Eight: Itrducti t Iferece Syllabus Objectives: 4.1 The studet will estimate ppulati parameters ad margis f errrs fr meas. 4.2 The studet will discuss the prperties f pit estimatrs,

More information

Author. Introduction. Author. o Asmir Tobudic. ISE 599 Computational Modeling of Expressive Performance

Author. Introduction. Author. o Asmir Tobudic. ISE 599 Computational Modeling of Expressive Performance ISE 599 Cmputatial Mdelig f Expressive Perfrmace Playig Mzart by Aalgy: Learig Multi-level Timig ad Dyamics Strategies by Gerhard Widmer ad Asmir Tbudic Preseted by Tsug-Ha (Rbert) Chiag April 5, 2006

More information

Study in Cylindrical Coordinates of the Heat Transfer Through a Tow Material-Thermal Impedance

Study in Cylindrical Coordinates of the Heat Transfer Through a Tow Material-Thermal Impedance Research ural f Applied Scieces, Egieerig ad echlgy (): 9-63, 3 ISSN: 4-749; e-issn: 4-7467 Maxwell Scietific Orgaiati, 3 Submitted: uly 4, Accepted: September 8, Published: May, 3 Study i Cylidrical Crdiates

More information

Control Systems. Controllability and Observability (Chapter 6)

Control Systems. Controllability and Observability (Chapter 6) 6.53 trl Systems trllaility ad Oservaility (hapter 6) Geeral Framewrk i State-Spae pprah Give a LTI system: x x u; y x (*) The system might e ustale r des t meet the required perfrmae spe. Hw a we imprve

More information

Review of Important Concepts

Review of Important Concepts Appedix 1 Review f Imprtat Ccepts I 1 AI.I Liear ad Matrix Algebra Imprtat results frm liear ad matrix algebra thery are reviewed i this secti. I the discussis t fllw it is assumed that the reader already

More information

MATH Midterm Examination Victor Matveev October 26, 2016

MATH Midterm Examination Victor Matveev October 26, 2016 MATH 33- Midterm Examiati Victr Matveev Octber 6, 6. (5pts, mi) Suppse f(x) equals si x the iterval < x < (=), ad is a eve peridic extesi f this fucti t the rest f the real lie. Fid the csie series fr

More information

x 2 x 3 x b 0, then a, b, c log x 1 log z log x log y 1 logb log a dy 4. dx As tangent is perpendicular to the x axis, slope

x 2 x 3 x b 0, then a, b, c log x 1 log z log x log y 1 logb log a dy 4. dx As tangent is perpendicular to the x axis, slope The agle betwee the tagets draw t the parabla y = frm the pit (-,) 5 9 6 Here give pit lies the directri, hece the agle betwee the tagets frm that pit right agle Ratig :EASY The umber f values f c such

More information

IJISET - International Journal of Innovative Science, Engineering & Technology, Vol. 2 Issue 12, December

IJISET - International Journal of Innovative Science, Engineering & Technology, Vol. 2 Issue 12, December IJISET - Iteratial Jural f Ivative Sciece, Egieerig & Techlgy, Vl Issue, December 5 wwwijisetcm ISSN 48 7968 Psirmal ad * Pararmal mpsiti Operatrs the Fc Space Abstract Dr N Sivamai Departmet f athematics,

More information

Function representation of a noncommutative uniform algebra

Function representation of a noncommutative uniform algebra Fucti represetati f a cmmutative uifrm algebra Krzysztf Jarsz Abstract. We cstruct a Gelfad type represetati f a real cmmutative Baach algebra A satisfyig f 2 = kfk 2, fr all f 2 A:. Itrducti A uifrm algebra

More information

The generation of successive approximation methods for Markov decision processes by using stopping times

The generation of successive approximation methods for Markov decision processes by using stopping times The geerati f successive apprximati methds fr Markv decisi prcesses by usig stppig times Citati fr published versi (APA): va Nue, J. A. E. E., & Wessels, J. (1976). The geerati f successive apprximati

More information

Preliminary Test Single Stage Shrinkage Estimator for the Scale Parameter of Gamma Distribution

Preliminary Test Single Stage Shrinkage Estimator for the Scale Parameter of Gamma Distribution America Jural f Mathematics ad Statistics, (3): 3-3 DOI:.593/j.ajms.3. Prelimiary Test Sigle Stage Shrikage Estimatr fr the Scale Parameter f Gamma Distributi Abbas Najim Salma,*, Aseel Hussei Ali, Mua

More information

Unifying the Derivations for. the Akaike and Corrected Akaike. Information Criteria. from Statistics & Probability Letters,

Unifying the Derivations for. the Akaike and Corrected Akaike. Information Criteria. from Statistics & Probability Letters, Uifyig the Derivatis fr the Akaike ad Crrected Akaike Ifrmati Criteria frm Statistics & Prbability Letters, Vlume 33, 1997, pages 201{208. by Jseph E. Cavaaugh Departmet f Statistics, Uiversity f Missuri,

More information

European Journal of Operational Research

European Journal of Operational Research Eurpea Jural f Operatial Research 232 (2014) 4 4 Ctets lists available at ScieceDirect Eurpea Jural f Operatial Research jural hmepage www.elsevier.cm/lcate/ejr Discrete Optimizati A aalytical cmparis

More information

On the Consistency of Slacks-based Measure-Max Model and Super-Slacks-based Measure Model

On the Consistency of Slacks-based Measure-Max Model and Super-Slacks-based Measure Model GRIPS Dicui Paper 6-24 O the Citecy Slack-baed Meaure-Max Mdel ad Super-Slack-baed Meaure Mdel Karu Te Nveber 206 Natial Graduate Ititute r Plicy Studie 7-22- Rppgi, Miat-ku, Tky, Japa 06-8677 O the Citecy

More information

Distributed Trajectory Generation for Cooperative Multi-Arm Robots via Virtual Force Interactions

Distributed Trajectory Generation for Cooperative Multi-Arm Robots via Virtual Force Interactions 862 IEEE TRANSACTIONS ON SYSTEMS, MAN, AND CYBERNETICS PART B: CYBERNETICS, VOL. 27, NO. 5, OCTOBER 1997 Distributed Trajectry Geerati fr Cperative Multi-Arm Rbts via Virtual Frce Iteractis Tshi Tsuji,

More information

Lecture 21: Signal Subspaces and Sparsity

Lecture 21: Signal Subspaces and Sparsity ECE 830 Fall 00 Statistical Sigal Prcessig istructr: R. Nwak Lecture : Sigal Subspaces ad Sparsity Sigal Subspaces ad Sparsity Recall the classical liear sigal mdel: X = H + w, w N(0, where S = H, is a

More information

A unified brittle fracture criterion for structures with sharp V-notches under mixed mode loading

A unified brittle fracture criterion for structures with sharp V-notches under mixed mode loading Jural f Mechaical Sciece ad Techlgy Jural f Mechaical Sciece ad Techlgy 22 (2008) 269~278 www.sprigerlik.cm/ctet/738-494x A uified brittle fracture criteri fr structures with sharp V-tches uder mixed mde

More information

MATHEMATICS 9740/01 Paper 1 14 Sep hours

MATHEMATICS 9740/01 Paper 1 14 Sep hours Cadidate Name: Class: JC PRELIMINARY EXAM Higher MATHEMATICS 9740/0 Paper 4 Sep 06 3 hurs Additial Materials: Cver page Aswer papers List f Frmulae (MF5) READ THESE INSTRUCTIONS FIRST Write yur full ame

More information

Markov processes and the Kolmogorov equations

Markov processes and the Kolmogorov equations Chapter 6 Markv prcesses ad the Klmgrv equatis 6. Stchastic Differetial Equatis Csider the stchastic differetial equati: dx(t) =a(t X(t)) dt + (t X(t)) db(t): (SDE) Here a(t x) ad (t x) are give fuctis,

More information

Frequency-Domain Study of Lock Range of Injection-Locked Non- Harmonic Oscillators

Frequency-Domain Study of Lock Range of Injection-Locked Non- Harmonic Oscillators 0 teratial Cferece mage Visi ad Cmputig CVC 0 PCST vl. 50 0 0 ACST Press Sigapre DO: 0.776/PCST.0.V50.6 Frequecy-Dmai Study f Lck Rage f jecti-lcked N- armic Oscillatrs Yushi Zhu ad Fei Yua Departmet f

More information

Claude Elysée Lobry Université de Nice, Faculté des Sciences, parc Valrose, NICE, France.

Claude Elysée Lobry Université de Nice, Faculté des Sciences, parc Valrose, NICE, France. CHAOS AND CELLULAR AUTOMATA Claude Elysée Lbry Uiversité de Nice, Faculté des Scieces, parc Valrse, 06000 NICE, Frace. Keywrds: Chas, bifurcati, cellularautmata, cmputersimulatis, dyamical system, ifectius

More information

Unit -2 THEORY OF DILUTE SOLUTIONS

Unit -2 THEORY OF DILUTE SOLUTIONS Uit - THEORY OF DILUTE SOLUTIONS 1) hat is sluti? : It is a hmgeus mixture f tw r mre cmpuds. ) hat is dilute sluti? : It is a sluti i which slute ccetrati is very less. 3) Give a example fr slid- slid

More information

5.80 Small-Molecule Spectroscopy and Dynamics

5.80 Small-Molecule Spectroscopy and Dynamics MIT OpeCurseWare http://cw.mit.edu 5.8 Small-Mlecule Spectrscpy ad Dyamics Fall 8 Fr ifrmati abut citig these materials r ur Terms f Use, visit: http://cw.mit.edu/terms. 5.8 Lecture #33 Fall, 8 Page f

More information

Chapter 1: Fundamentals

Chapter 1: Fundamentals Chapter 1: Fudametals 1.1 Real Numbers Irratial umbers are real umbers that cat be expressed as ratis f itegers. That such umbers exist was a prfud embarrassmet t the Pythagrea brtherhd, ad they are said

More information

Design and Implementation of Cosine Transforms Employing a CORDIC Processor

Design and Implementation of Cosine Transforms Employing a CORDIC Processor C16 1 Desig ad Implemetati f Csie Trasfrms Emplyig a CORDIC Prcessr Sharaf El-Di El-Nahas, Ammar Mttie Al Hsaiy, Magdy M. Saeb Arab Academy fr Sciece ad Techlgy, Schl f Egieerig, Alexadria, EGYPT ABSTRACT

More information

A Simplified Nonlinear Generalized Maxwell Model for Predicting the Time Dependent Behavior of Viscoelastic Materials

A Simplified Nonlinear Generalized Maxwell Model for Predicting the Time Dependent Behavior of Viscoelastic Materials Wrld Jural f Mechaics, 20,, 58-67 di:0.4236/wj.20.302 Published Olie Jue 20 (http://www.scirp.rg/jural/wj) A Siplified Nliear Geeralized Maxwell Mdel fr Predictig the Tie Depedet Behavir f Viscelastic

More information

RMO Sample Paper 1 Solutions :

RMO Sample Paper 1 Solutions : RMO Sample Paper Slutis :. The umber f arragemets withut ay restricti = 9! 3!3!3! The umber f arragemets with ly e set f the csecutive 3 letters = The umber f arragemets with ly tw sets f the csecutive

More information

COMPUTING CONFIDENCE INTERVALS FOR OUTPUT ORIENTED DEA MODELS: AN APPLICATION TO AGRICULTURAL RESEARCH IN BRAZIL

COMPUTING CONFIDENCE INTERVALS FOR OUTPUT ORIENTED DEA MODELS: AN APPLICATION TO AGRICULTURAL RESEARCH IN BRAZIL COMPUTING CONFIDENCE INTERVALS FOR OUTPUT ORIENTED DEA MODELS: AN APPLICATION TO AGRICULTURAL RESEARCH IN BRAZIL Gerald da Silva e Suza Miria Oliveira de Suza Eliae Gçalves Gmes Brazilia Agricultural Research

More information

Full algebra of generalized functions and non-standard asymptotic analysis

Full algebra of generalized functions and non-standard asymptotic analysis Full algebra f geeralized fuctis ad -stadard asympttic aalysis Tdr D. Tdrv Has Veraeve Abstract We cstruct a algebra f geeralized fuctis edwed with a caical embeddig f the space f Schwartz distributis.

More information

On the affine nonlinearity in circuit theory

On the affine nonlinearity in circuit theory O the affie liearity i circuit thery Emauel Gluski The Kieret Cllege the Sea f Galilee; ad Ort Braude Cllege (Carmiel), Israel. gluski@ee.bgu.ac.il; http://www.ee.bgu.ac.il/~gluski/ E. Gluski, O the affie

More information

Every gas consists of a large number of small particles called molecules moving with very high velocities in all possible directions.

Every gas consists of a large number of small particles called molecules moving with very high velocities in all possible directions. Kietic thery f gases ( Kietic thery was develped by Berlli, Jle, Clasis, axwell ad Bltzma etc. ad represets dyamic particle r micrscpic mdel fr differet gases sice it thrws light the behir f the particles

More information

This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and

This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and This article appeared i a jural published by Elsevier The attached cpy is furished t the authr fr iteral -cmmercial research ad educati use, icludig fr istructi at the authrs istituti ad sharig with clleagues

More information

6.867 Machine learning, lecture 14 (Jaakkola)

6.867 Machine learning, lecture 14 (Jaakkola) 6.867 Machie learig, lecture 14 (Jaakkla) 1 Lecture tpics: argi ad geeralizati liear classifiers esebles iture dels Margi ad geeralizati: liear classifiers As we icrease the uber f data pits, ay set f

More information

Solutions to Midterm II. of the following equation consistent with the boundary condition stated u. y u x y

Solutions to Midterm II. of the following equation consistent with the boundary condition stated u. y u x y Sltis t Midterm II Prblem : (pts) Fid the mst geeral slti ( f the fllwig eqati csistet with the bdary cditi stated y 3 y the lie y () Slti : Sice the system () is liear the slti is give as a sperpsiti

More information

WEST VIRGINIA UNIVERSITY

WEST VIRGINIA UNIVERSITY WEST VIRGINIA UNIVERSITY PLASMA PHYSICS GROUP INTERNAL REPORT PL - 045 Mea Optical epth ad Optical Escape Factr fr Helium Trasitis i Helic Plasmas R.F. Bivi Nvember 000 Revised March 00 TABLE OF CONTENT.0

More information

UNIVERSITY OF TECHNOLOGY. Department of Mathematics PROBABILITY THEORY, STATISTICS AND OPERATIONS RESEARCH GROUP. Memorandum COSOR 76-10

UNIVERSITY OF TECHNOLOGY. Department of Mathematics PROBABILITY THEORY, STATISTICS AND OPERATIONS RESEARCH GROUP. Memorandum COSOR 76-10 EI~~HOVEN UNIVERSITY OF TECHNOLOGY Departmet f Mathematics PROBABILITY THEORY, STATISTICS AND OPERATIONS RESEARCH GROUP Memradum COSOR 76-10 O a class f embedded Markv prcesses ad recurrece by F.H. Sims

More information

Rates and Mechanisms of Chemical Reactions

Rates and Mechanisms of Chemical Reactions Rates ad Mechaisms f Chemical Reactis Why sme rxs prceed very fast ad thers require days, mths r eve years t prduce a detectable amt f prduct? H (g) + F (g) HF (g) (very fast) 3 H (g) + N (g) NH 3 (g)

More information

Review for cumulative test

Review for cumulative test Hrs Math 3 review prblems Jauary, 01 cumulative: Chapters 1- page 1 Review fr cumulative test O Mday, Jauary 7, Hrs Math 3 will have a curse-wide cumulative test cverig Chapters 1-. Yu ca expect the test

More information

Energy xxx (2011) 1e10. Contents lists available at ScienceDirect. Energy. journal homepage:

Energy xxx (2011) 1e10. Contents lists available at ScienceDirect. Energy. journal homepage: Eergy xxx (2011) 1e10 Ctets lists available at ScieceDirect Eergy jural hmepage: www.elsevier.cm/lcate/eergy Multi-bjective ptimizati f HVAC system with a evlutiary cmputati algrithm Adrew Kusiak *, Fa

More information

Pipe Networks - Hardy Cross Method Page 1. Pipe Networks

Pipe Networks - Hardy Cross Method Page 1. Pipe Networks Pie Netwrks - Hardy Crss etd Page Pie Netwrks Itrducti A ie etwrk is a itercected set f ies likig e r mre surces t e r mre demad (delivery) its, ad ca ivlve ay umber f ies i series, bracig ies, ad arallel

More information

An Investigation of Stratified Jackknife Estimators Using Simulated Establishment Data Under an Unequal Probability Sample Design

An Investigation of Stratified Jackknife Estimators Using Simulated Establishment Data Under an Unequal Probability Sample Design Secti Survey Research Methds SM 9 A Ivestigati f Stratified ackkife Estimatrs Usig Simulated Establishmet Data Uder a Uequal Prbability Sample Desig Abstract Plip Steel, Victria McNerey, h Slata Csiderig

More information

Pattern Recognition 2014 Support Vector Machines

Pattern Recognition 2014 Support Vector Machines Pattern Recgnitin 2014 Supprt Vectr Machines Ad Feelders Universiteit Utrecht Ad Feelders ( Universiteit Utrecht ) Pattern Recgnitin 1 / 55 Overview 1 Separable Case 2 Kernel Functins 3 Allwing Errrs (Sft

More information

Dynamic Response of Second Order Mechanical Systems with Viscous Dissipation forces

Dynamic Response of Second Order Mechanical Systems with Viscous Dissipation forces Hadut #a (pp. 1-39) Dyamic Respse f Secd Order Mechaical Systems with Viscus Dissipati frces d X d X + + = ext() t M D K X F dt dt Free Respse t iitial cditis ad F (t) = 0, Uderdamped, Critically Damped

More information

Copyright 1978, by the author(s). All rights reserved.

Copyright 1978, by the author(s). All rights reserved. Cpyright 1978, by the authr(s). All rights reserved. Permissi t make digital r hard cpies f all r part f this wrk fr persal r classrm use is grated withut fee prvided that cpies are t made r distributed

More information

Performance Impact of E-Business Initiatives on the US Retail Industry. Yao Chen, Luvai Motiwalla and M. Riaz Khan

Performance Impact of E-Business Initiatives on the US Retail Industry. Yao Chen, Luvai Motiwalla and M. Riaz Khan Perfrmace Impact f E-Busiess Iitiatives the US Retail Idustr Ya Che, Luvai Mtiwalla ad M. Riaz Kha Cllege f Maagemet Uiversit f Massachusetts Lwell, MA 0845 USA Email: Ya_Che@uml.edu Jauar 2003 Perfrmace

More information

Strategy in practice: a quantitative approach to target setting

Strategy in practice: a quantitative approach to target setting MPRA Muich Peral RePEc Archive Strategy i practice: a quatitative apprach t target ettig Iree Fafaliu ad Paagiti Zervpul Uiverity f Piraeu, Ope Uiverity f Cypru 4. Jauary 2014 Olie at http://mpra.ub.ui-mueche.de/54054/

More information

Chapter 4. Problem Solutions

Chapter 4. Problem Solutions Chapter 4. Prblem Slutis. The great majrity f alpha particles pass thrugh gases ad thi metal fils with deflectis. T what cclusi abut atmic structure des this bservati lead? The fact that mst particles

More information

Declarative approach to cyclic steady state space refinement: periodic process scheduling

Declarative approach to cyclic steady state space refinement: periodic process scheduling It J Adv Mauf Techl DOI 10.1007/s00170-013-4760-0 ORIGINAL ARTICLE Declarative apprach t cyclic steady state space refiemet: peridic prcess schedulig Grzegrz Bcewicz Zbigiew A. Baaszak Received: 16 April

More information

8.0 Negative Bias Temperature Instability (NBTI)

8.0 Negative Bias Temperature Instability (NBTI) EE650R: Reliability Physics f Naelectric Devices Lecture 8: Negative Bias Temerature Istability Date: Se 27 2006 Class Ntes: Vijay Rawat Reviewed by: Saakshi Gagwal 8.0 Negative Bias Temerature Istability

More information

Identical Particles. We would like to move from the quantum theory of hydrogen to that for the rest of the periodic table

Identical Particles. We would like to move from the quantum theory of hydrogen to that for the rest of the periodic table We wuld like t ve fr the quatu thery f hydrge t that fr the rest f the peridic table Oe electr at t ultielectr ats This is cplicated by the iteracti f the electrs with each ther ad by the fact that the

More information

Abstract: The asympttically ptimal hypthesis testig prblem with the geeral surces as the ull ad alterative hyptheses is studied uder expetial-type err

Abstract: The asympttically ptimal hypthesis testig prblem with the geeral surces as the ull ad alterative hyptheses is studied uder expetial-type err Hypthesis Testig with the Geeral Surce y Te Su HAN z April 26, 2000 y This paper is a exteded ad revised versi f Sectis 4.4 4.7 i Chapter 4 f the Japaese bk f Ha [8]. z Te Su Ha is with the Graduate Schl

More information

Experimental and Theoretical Investigations of PAN Molecular Weight Increase in Precipitation Polymerization as a Function of H 2 O/DMSO Ratio

Experimental and Theoretical Investigations of PAN Molecular Weight Increase in Precipitation Polymerization as a Function of H 2 O/DMSO Ratio Carb Letters Vl., N. March 00 pp. -7 Experimetal ad Theretical Ivestigatis f PAN Mlecular Weight Icrease i Precipitati Plymerizati as a ucti f H O/DMSO Rati Jig Zhag, egjig Bu, Ygqiag Dai, Liwei Xue, Zhixia

More information

Analysis of pressure wave dynamics in fuel rail system

Analysis of pressure wave dynamics in fuel rail system It. Jl. f Multiphysics Vlume Number 3 8 3 Aalysis f pressure wave dyamics i fuel rail system Basem Alzahabi ad Keith Schulz Ketterig Uiversity Siemes Autmtive ABSTRACT A mdel f a amplified cmm rail fuel

More information

ANALOG FILTERS. C. Sauriol. Algonquin College Ottawa, Ontario

ANALOG FILTERS. C. Sauriol. Algonquin College Ottawa, Ontario LOG ILT By. auril lgqui llege Ottawa, Otari ev. March 4, 003 TBL O OTT alg ilters TIO PI ILT. irst-rder lw-pass filter- -4. irst-rder high-pass filter- 4-6 3. ecd-rder lw-pass filter- 6-4. ecd-rder bad-pass

More information

E o and the equilibrium constant, K

E o and the equilibrium constant, K lectrchemical measuremets (Ch -5 t 6). T state the relati betwee ad K. (D x -b, -). Frm galvaic cell vltage measuremet (a) K sp (D xercise -8, -) (b) K sp ad γ (D xercise -9) (c) K a (D xercise -G, -6)

More information

ON FREE RING EXTENSIONS OF DEGREE N

ON FREE RING EXTENSIONS OF DEGREE N I terat. J. Math. & Mah. Sci. Vl. 4 N. 4 (1981) 703-709 703 ON FREE RING EXTENSIONS OF DEGREE N GEORGE SZETO Mathematics Departmet Bradley Uiversity Peria, Illiis 61625 U.S.A. (Received Jue 25, 1980) ABSTRACT.

More information

Electrostatics. . where,.(1.1) Maxwell Eqn. Total Charge. Two point charges r 12 distance apart in space

Electrostatics. . where,.(1.1) Maxwell Eqn. Total Charge. Two point charges r 12 distance apart in space Maxwell Eq. E ρ Electrstatics e. where,.(.) first term is the permittivity i vacuum 8.854x0 C /Nm secd term is electrical field stregth, frce/charge, v/m r N/C third term is the charge desity, C/m 3 E

More information

ESWW-2. Israeli semi-underground great plastic scintillation multidirectional muon telescope (ISRAMUTE) for space weather monitoring and forecasting

ESWW-2. Israeli semi-underground great plastic scintillation multidirectional muon telescope (ISRAMUTE) for space weather monitoring and forecasting ESWW-2 Israeli semi-udergrud great plastic scitillati multidirectial mu telescpe (ISRAMUTE) fr space weather mitrig ad frecastig L.I. Drma a,b, L.A. Pustil'ik a, A. Sterlieb a, I.G. Zukerma a (a) Israel

More information

Roel Jongeneel

Roel Jongeneel A Aalysis f the Impact f Alterative EU Dairy Plicies the Size Distributi f Dutch Dairy Farms: a Ifrmati Based Apprach t the N- Statiary Markv Chai Mdel Rel Jgeeel E-mail: rel.jgeeel@alg.aae.wag-ur.l Paper

More information

Probabilistic linguistic TODIM approach for multiple attribute decision-making

Probabilistic linguistic TODIM approach for multiple attribute decision-making Graul. Cmput. (07) : 4 DOI 0.007/s4066-07-0047-4 ORIGINAL PAPER Prbabilistic liguistic TODIM apprach fr multiple attribute decisi-makig Peide Liu Xili Yu Received: 9 April 07 / Accepted: 5 July 07 / Published

More information

Recovery of Third Order Tensors via Convex Optimization

Recovery of Third Order Tensors via Convex Optimization Recvery f Third Order Tesrs via Cvex Optimizati Hlger Rauhut RWTH Aache Uiversity Lehrstuhl C für Mathematik (Aalysis) Ptdriesch 10 5056 Aache Germay Email: rauhut@mathcrwth-aachede Željka Stjaac RWTH

More information

The generalized marginal rate of substitution

The generalized marginal rate of substitution Jural f Mathematical Ecmics 31 1999 553 560 The geeralized margial rate f substituti M Besada, C Vazuez ) Facultade de Ecmicas, UiÕersidade de Vig, Aptd 874, 3600 Vig, Spai Received 31 May 1995; accepted

More information

Chapter 5. Root Locus Techniques

Chapter 5. Root Locus Techniques Chapter 5 Rt Lcu Techique Itrducti Sytem perfrmace ad tability dt determied dby cled-lp l ple Typical cled-lp feedback ctrl ytem G Ope-lp TF KG H Zer -, - Ple 0, -, - K Lcati f ple eaily fud Variati f

More information

Support-Vector Machines

Support-Vector Machines Supprt-Vectr Machines Intrductin Supprt vectr machine is a linear machine with sme very nice prperties. Haykin chapter 6. See Alpaydin chapter 13 fr similar cntent. Nte: Part f this lecture drew material

More information

The Molecular Diffusion of Heat and Mass from Two Spheres

The Molecular Diffusion of Heat and Mass from Two Spheres Iteratial Jural f Mder Studies i Mechaical Egieerig (IJMSME) Vlume 4, Issue 1, 018, PP 4-8 ISSN 454-9711 (Olie) DOI: http://dx.di.rg/10.0431/454-9711.0401004 www.arcjurals.rg The Mlecular Diffusi f Heat

More information