Improving the Performance of PCA-Based Chiller Sensor Fault Detection by Sensitivity Analysis for the Training Data Set
|
|
- Edgar Garrison
- 5 years ago
- Views:
Transcription
1 Purdue Uversty Purdue e-pubs Iteratal gh Perfrmace Buldgs Cferece Schl f Mechacal Egeerg 016 Imprvg the Perfrmace f PCA-Based Chller Sesr Fault Detect by Sestvty Aalyss fr the rag Data Set Yupeg u Wuha Buzess Uversty, Cha, Peple's Republc f, yupeghu@hust.edu.c Ja Lu Wuha Buzess Uversty, Cha, Peple's Republc f, @qq.cm L Zhu Wuha Buzess Uversty, Cha, Peple's Republc f, @qq.cm Yag Lu Wuha Buzess Uversty, Cha, Peple's Republc f, @qq.cm Qglg Qu Wuha Buzess Uversty, Cha, Peple's Republc f, @qq.cm Fllw ths ad addtal wrks at: u, Yupeg; Lu, Ja; Zhu, L; Lu, Yag; ad Qu, Qglg, "Imprvg the Perfrmace f PCA-Based Chller Sesr Fault Detect by Sestvty Aalyss fr the rag Data Set" 016). Iteratal gh Perfrmace Buldgs Cferece. Paper hs dcumet has bee made avalable thrugh Purdue e-pubs, a servce f the Purdue Uversty Lbrares. Please ctact epubs@purdue.edu fr addtal frmat. Cmplete prceedgs may be acqured prt ad CD-ROM drectly frm the Ray W. errck Labratres at errck/evets/rderlt.html
2 318, Page 1 Imprvg the Perfrmace f PCA-Based Chller Sesr Fault Detect by Sestvty Aalyss fr the rag Data Set Yupeg U*, Ja LIU, L ZOU, Yag LIU, Qglg QIU Wuha Busess Uversty, Departmet f Buldg Evrmet ad Eergy Egeerg, , Wuha, ube, PR Cha * Crrespdg Authr : , Yupegu@hust.edu.c ABSRAC A mprved apprach f fault detect fr chller sesrs s preseted based the sestvty aalyss fr the rgal data set used t tra the Prcpal Cmpet Aalyss PCA) mdel. Sesr faults are evtable due t the agg, evrmet, lcat ad s. Meawhle, because f the wde rage f peratal cdts, the fault f a certa sesr s very dffcult t be drectly detected by ts w hstrcal data. PCA s a multvarate data-based statstcal aalyss methd ad t s very useful fr the sesr fault detect VAC&R. he udetectable ze f a certa sesr by Q-statstc s derved frm the deft f Q-statstc whch s usually emplyed as a budary t detect the sesr fault stuat. Due t the smlar style betwee Q-statstc ad awks, the udetectable ze by awks s als btaed. Udetectable ze s a predctve dex t dcate the detectablty f dfferet sesrs by dfferet statstcs. Sce udetectable ze s the character f the rgal trag data set, t ca dcate the qualty fr the selected trag data. Oe feld data set s emplyed t valdate the preseted apprach. Results shw that the udetectable ze f a certa sesr by Q-statstc s qute dfferet frm that by awks. herefre, the udetectable ze ca be used t mprvg the perfrmace f PCA-based chller sesr fault detect by chsg dfferet fault detect statstcs wth less udetectable ze fr dfferet sesr. 1. INRODUCION Due t lg term perat ad severely wrkg evrmet, sesr faults are evtable VAC&R. here are much dsadvatage because f sesr fault, cludg effectve ctrl, usafe perat, ureasable eergy csumpt ad s Lee ad Yk, 010; Y et al., 011). Fr eergy savg ad cservat, researches sesr fault detect, dagss ad erreus sesr data recstruct FDDR) fr VAC&R system have bee pad mre attet t the last decade. Usually, the mdel-based methds ad the data-drve methds are the tw typcal classes f FDDR methds. Ay faulty sesr cat be easly detfed ust ly frm the hstrcal data f ts w. hus, varus multdmesal data-based methds have bee trduced t the FDDR f VAC&R system the recet years, such as fuzzy ferece systemskcygt, 015), data fussu et al., 010), eural etwrkdu et al., 014; Lee et al., 004), supprt vectr machea et al., 011), prcpal cmpet aalyssl et al., 016), fsher dscrmat aalyssdu et al., 007), Bayesa etwrkzha et al., 015), etc. Recet years, prcpal cmpet aalyss PCA) ärdle ad Smar, 007; Jacks, 1991), a multvarate statstcal aalyss methd, was preseted the sesr FDDR, cludg the whle systemwag et al., 010), AUL ad We, 014; Xa et al., 009), VAVDu et al., 009), chllerche ad La, 009; u et al., 016; Xu et al., 008) ad s. By the dfferet assgmet f sesrs r the cmbat wth ther algrthms, PCA-based appraches were successfully appled t sesr FDDR fr chller. May researchers were dedcated applyg vel data-drve methds t the sesr FDDR f VAC&R system. Fr ay data-drve methd, the aalyss results hghly deped the character f the trag data set. Because the felded data reles samplg terval, samplg lcat, measuremet prcples, ad s, the qualty f trag data s much wrse. wever, rare wrk was reprted hw t predct the qualty f the trag data s as t ehace the FDDR results detal. I ths paper, the udetectable ze fr each sesr assged PCA mdel s 4 th Iteratal gh Perfrmace Buldgs Cferece at Purdue, July 11-14, 016
3 318, Page preseted t evaluate the detectablty fr chller sesr fault. It s derved frm the trag data ad the deft f statstcs emplyed as fault detect budary. he udetectable ze fr each sesr ca be used t evaluate the fault detect ablty ad relablty wth clear physcal r thermdyamc meags. A felded data set f a real screw chller was emplyed t valdate the detectablty f dfferet statstcs detal.. PCA-BASED SENSOR FDDR.1 PCA-based sesr FDDR 0 m I PCA methd, the rgal data matrx X R usually cssts f m samples rws) ad prcess varables 0 m clums) btaed frm the feld measuremets. he trag data X, whch s cssted f the rgal measured data, s trasferred t a rmalzed matrx,, X x1 x m wth zer mea ad ut varace due t egeerg uts ad rders f magtude. After the egevalue decmpst f the cvarace matrx, R X X 1), ay rmalzed samples x ca be expressed as x xˆ x 1) where ˆx, the estmat f x, s the prect vectr f x t the PC subspace, ad x, the resdual f x, s the prect vectrs f x t the Resdual subspace. A cmm FDDR strategy fr sesr fault based PCA s llustrated as Fgure 1. Its detaled structure ca be referred referece u et al., 01). It eeds t emphasze that the rgal peratal data used t tra PCA mdel s cluded wth may utlers evtably due t measuremet errrs, hardware falure ad s. he am f PCA mdelg s t establsh a fault budary t detect whether there s a faulty sesr system r t. Fgure 1: A cmm FDDR strategy fr sesr fault based PCA Several statstcs ca be emplyed as the budary t detect sesr fault, such as Q, awks ad s. Whe the value f statstcs fr the tested sample s greater tha the budary, the fault ca be detected successfully. herefre, the belw equats mea the sesr fault cat be detected. Q Q ) 3) k; Where, Q α s the threshld f Q statstc ad k; s the threshld f awks. Udetectable Ze by Q-statstc If there s a faulty sesr, the measuremet data f ths sesr make the value f sme statstc s greater tha the threshld. Assumg the certa faulty sesr s the th sesr, there must be a par f lmted upsde ad dwsde f th sesr measuremet data, whch ca ust satsfy the Equat 1) r Equat ). Whe the measuremet data f the th sesr s utsde the par f lmtats, the fault ca be detected. Frm the threshld ad the ther sesrs. 4 th Iteratal gh Perfrmace Buldgs Cferece at Purdue, July 11-14, 016
4 318, Page 3 measuremet data, we derved the calculat f the lmtats fr the th sesr t t be detected. Obvusly, the par f lmtats s a predctve ze t demstrate the fault detectablty fr the th sesr. Assumg the th sesr s the faulty e, x, the th etry f x, s the erreus measuremet value. e, the th etry f x, ca be rewrtte as 1 RS RS RS RS RS k k k k,: k1 k 1 4) e y x y x y x y x Y x Where, y s the th etry f the th rw f RS drect f the erreus ser ad be wrtte as RS Y. Y RS s the prect matrx f RS. s used t dcate the ,, 1 1,, 5) herefre, Q-statstc ca be derved as Q y x y Y x x Y x RS RS RS RS ),: )),: ) ) Due t the th sesr s faulty e, ts Q-statstc wll satsfy the fllwg equat 7) y ) x y Y x)) x Y x) Q 0 RS RS RS RS,:,: ) Frm the style f Equat, t s a e-varable quadratc equaltes wth the frm f ax bx c>0. Where, a 1 1 y ) RS, RS RS b y Y,: x), 8) RS c Y,: x) -Q 1 he par f sluts f Equat 7), x,m ad x,max, are the lmtats fr the rmalzed sesr fault budary. A ze, 0 0 [ x, x ], ca be btaed by de-rmalzg. If the rgal measured data 0 x s utsde f 0 0 [ x, x ], we,m,max ca easly fd the faulty sesr. herefre, the area, 0 0 [ x,m, x,max ], ca be defed as the udetectable ze f the th sesr by Q-statstc. he udetectable ze ca be emplyed as the dex t evaluate the sesr fault detectablty..3 Udetectable Ze by Smlar wth the deft f Q-statstc, the awks ca be rewrtte as,m,max y x y Y x x Y x ),: )),: ) ) Where, 1/ Y P s the prect matrx f awks. herefre, the udetectable ze f the th sesr by k1, m awks ca be the sluts wth the style f ax bx c>0, where 4 th Iteratal gh Perfrmace Buldgs Cferece at Purdue, July 11-14, 016
5 318, Page 4 a 1 y ),: ) 1 b y Y x c Y,: x) -k ; 1 10) he udetectable ze by awks s qute dfferet t that by Q-statstc. herefre, the cmpared results ca dcate the dfferet sestvty fr these tw dfferet statstcs..4 PCA mdel fr a water-cled chller Frm the csderat f the eergy balace prcple, there are eght mprtat sesrs the water-cled chller ad ts ctrl lgc. he PCA mdel f a typcal water-cled chller s X M M W V 11) Where, ad are the temperature sesrs f let de ad utlet de f evapratr, respectvely. ad are the cdeser-water system let de temperature ad utlet de temperature, respectvely. M s the water flwrate f chlled-water system ad M s the water flwrate f cdeser-water system, respectvely. W s the electrcal pwer. V s the pst f the slde valve t dcate the mass flwrate f the refrgerat t the screw cmpressr. 3. VALIDAION 3.1 Cases study A felded data u et al., 01; u et al., 016) f a water-cled screw chller were used t valdate the sestvty f Q ad awks fr dfferet sesrs. he udetectable zes f dfferet sesrs by a same trag data set were vestgated. he results f sesr fault detect were used t valdate the predctve ablty f udetectable ze fr the faulty sesr. CASE I: wth -1.5 bas fault Udetectable ze f fr the trag data set s llustrated the Fgure. Fgure : Udetectable ze f fr the trag data set 4 th Iteratal gh Perfrmace Buldgs Cferece at Purdue, July 11-14, 016
6 318, Page 5 he up lmts f udetectable ze by Q s almst equal t that by awks, as well as the dw lmts f tw statstcs. Obvusly, the fault detectablty f by Q s equal t the ablty by awks. he udetectable ze f by Q s ±1.39, whle that by awks s ±0.67.here are ust the frmer 0 samples shw the hrztal axs rder t make the fgure clear. A bas fault wth -1.5 was trduced t t test the predctve ablty. he fault detect results by Q ad are llustrated Fgure 3 a) ad b). All the Q-statstcs values f tested samples are greater tha the Q α ad the detect effcecy f sesr fault by Q s 100 %. Meawhle, the -1.5 bas fault f s cmpletely detected by awks. herefre, the detectablty dcated by the udetectable ze f s accrdg t the fault detect results f tested samples f. he udetectable ze successfully predcted the fault detect results. CASE II: wth -.0 bas fault a) b) Fgure 3: Fault detect fr wth -1.5 bas fault: a) by Q b) by Udetectable ze f fr the trag data set s llustrated the Fgure 4. Fgure 4: Udetectable ze f fr the trag data set 4 th Iteratal gh Perfrmace Buldgs Cferece at Purdue, July 11-14, 016
7 Ulke t the results f 318, Page 6, the up ad dw lmts f udetectable ze by Q s qute dfferet t that by awks. he udetectable ze by Q s much greater tha that by awks. he udetectable ze f by Q s ver ±6.0, whle that by awks s less tha ±1. Csequetly, the detectablty f fault by awks s much better tha the ablty by Q. he fault detect results by Q ad are llustrated Fgure 5 ad Fgure 6, respectvely, whe a.0 bas fault was trduced t. here are ly 1% f Q-statstcs values the tested samples are greater tha the Q α. It meas that the fault detect by Q dd t wrkg. Meawhle, the.0 bas fault f s cmpletely detected by awks. Due the trduced fault level,.0, s less tha the udetectable ze by Q, ±6.0, the detectablty by Q s cmpletely wrse tha that by awks. herefre, t s better that the awks s emplyed t detect the fault. Fgure 5: Fault detect fr wth.0 bas fault by Q Fgure 6: Fault detect fr wth.0 bas fault by 4 th Iteratal gh Perfrmace Buldgs Cferece at Purdue, July 11-14, 016
8 318, Page 7 CASE III: M wth +10% bas fault he predctve results, udetectable ze, f M fr the trag data set s llustrated the Fgure 7. Meawhle the fault detect results f M by Q ad are llustrated Fgure 8. he udetectable ze by Q s early equal t that by awks, s the fault detect results f M by Q are accrdace wth that by. Fgure 7: Udetectable ze f M fr the trag data set a) Fgure 8: Fault detect fr M wth +10% bas fault: a) by Q ad b) by 3. Summary he udetectable zes fr all sesrs the PCA mdel by Q ad by are summarzed the able 1. he detect abltes fr dfferet sesr by Q ad by are qute dfferet. At the step fr chsg the ptmal statstcs t bta the fault budary, the udetectable ze ca drectly predct the detect ablty fr the sesr by a certa statstcs, such as Q r. herefre, the perfrmace f PCA-based sesr Fault detect ca be mprved by chsg the statstcs wth hgher fault detect effcecy by the sestvty aalyss fr the certa trag data set. b) 4 th Iteratal gh Perfrmace Buldgs Cferece at Purdue, July 11-14, 016
9 318, Page 8 able 1: Summary fr all sesrs Udetectable zes by Q ad by Sesr Ut Udetectable ze by Q Udetectable ze by M l/m M W l/m kw M ref %.78 ±1.39).47 ±1.4) 6.8 ±3.14) ±5.89) 1.77 ±6.39) 7.18 ±3.59) ±68.9) ±0.75) 1.74 ±0.67) 1.68 ±0.84) 6.35 ±3.18) 1.66 ±0.84) 1.61 ±0.81) 5.30 ±.65) ±34.04) 0.33 ±10.17) here s a mprtat pt llustrated by the case study ad the summary. he slut f udetectable ze s derved frm the matrx calculat prcess f dfferet statstcs. he results f udetectable ze ly rely the rgal trag data. herefre, the udetectable ze demstrates the rgal feature f the trag data. 4. CONCLUSIONS Sesr fault detect, dagss ad erreus data recstruct s the fudametal wrk fr the thermdyamc fault slat, the ptmal ctrl, the safety perat ad s. I ths paper, a evaluat dex, udetectable ze, s preseted t predct the detectablty f sesr fault s as t mprve the perfrmace f sesr fault detect. Dfferet calculat algrthm s derved t bta the udetectable ze by dfferet statstcs. Udetectable ze ca be emplyed as a dex t predct ad t evaluate the detectablty f sesr fault by sme statstcs fr a certa trag data set. It ca be used t chse the ptmal statstcs f fault detect fr each sesr. Frm the evaluat f detectablty fr each sesr by dfferet statstcs, the le sesr fault detect ca be mre flexble by chsg the mst sestve statstcs. herefre, the detect effcecy ca be prmted by the predct f udetectable ze. NOMENCLAURE he meclature shuld be lcated at the ed f the text usg the fllwg frmat: temperature ) M water flw rate l/m) W chller electrcal-pwer put kw) V pst f the slde valve -) X rgal matrx -) X 0 rmalzed rgal matrx -) R cvarace matrx -) U ege vectr matrx -) VE varace explaed -) CV cumulatve ctrbut f varace -) FDD fault detect ad dagss -) FDDR fault detect, dagss ad recstruct -) VAC&R heatg, vetlatg, ar-cdtg ad refrgerat -) 4 th Iteratal gh Perfrmace Buldgs Cferece at Purdue, July 11-14, 016
10 318, Page 9 PC Prcpal Cmpet -) PCA Prcpal Cmpet Aalyss -) SPCA Prcpal Cmpet Aalyss wth a statstcal data-cleag -) P PC subspace prect matrx -) P Resdual subspace prect matrx -) Q α threshld f the Q-statstc -) x a sample -) ˆx estmate f a sample -) x resdual f a sample -) x recstruct f a sample -) rc x the th etry f x -) e the th etry f e -) Greek letters μ σ λ 1,, λ Superscrpt Subscrpt mea stadard devat egevalues let de utlet de chlled-water system cdeser-water system REFERENCES Du, Z. M., J, X. Q. & Wu, L. Z. 007). PCA-FDA-Based Fault Dagss fr Sesrs VAV Systems. VAC&R Research, 13), Du, Z. M., J, X. Q. & Yag, X. B. 009). A rbt fault dagstc tl fr flw rate sesrs ar dampers ad VAV termals. Eergy ad Buldgs, 413), Du, Z., Fa, B., Ch, J. & J, X. 014). Sesr fault detect ad ts effcecy aalyss ar hadlg ut usg the cmbed eural etwrks. Eergy ad Buldgs, 70), a,., Gu, B., Kag, J. & L, Z. R. 011). Study a hybrd SVM mdel fr chller FDD applcats. Appled hermal Egeerg, 314), ärdle, W. & Smar, L. 007). Appled Multvarate Statstcal Aalyss Secd ed.). New Yrk: Sprger Berl edelberg. u, Y., Che,., Xe, J., Yag, X. & Zhu, C. 01). Chller sesr fault detect usg a self-adaptve Prcpal Cmpet Aalyss methd. Eergy ad Buldgs, 54, Jacks, J. E. 1991). A User's Gude Prcpal Cmpets Frst ed.). New Yrk: Jh Wley & Ss, Ic. Kcygt, N. 015). Fault ad sesr errr dagstc strateges fr a vapr cmpress refrgerat system by usg fuzzy ferece systems ad artfcal eural etwrk. Iteratal Jural f Refrgerat, 500), Lee, S.. & Yk, F. W.. 010). A study the eergy pealty f varus ar-sde system faults buldgs. Eergy ad Buldgs, 41), -10. Lee, W., use, J. M. & Kyg, N. 004). Subsystem level fault dagss f a buldg's ar-hadlg ut usg geeral regress eural etwrks. Appled Eergy, 77), L, G., u, Y., Che,., She, L., L,., u, M., Lu, J. & Su, K. 016). A mprved fault detect methd fr cpet cetrfugal chller faults usg the PCA-R-SVDD algrthm. Eergy ad Buldgs, 116, L, S. & We, J. 014). A mdel-based fault detect ad dagstc methdlgy based PCA methd ad wavelet trasfrm. Eergy ad Buldgs, 68, Part A0), Su, Y. J., Wag, S. W. & uag, G. S. 010). Ole sesr fault dagss fr rbust chller sequecg ctrl. 4 th Iteratal gh Perfrmace Buldgs Cferece at Purdue, July 11-14, 016
11 318, Page 10 Iteratal Jural f hermal Sceces, 493), Wag, S. W., Zhu, Q. & Xa, F. 010). A system-level fault detect ad dagss strategy fr VAC systems vlvg sesr faults. Eergy ad Buldgs, 44), Xa, F., Wag, S. W., Xu, X.. & Ge, G. M. 009). A slat ehaced PCA methd wth expert-based multvarate decuplg fr sesr FDD ar-cdtg systems. Appled hermal Egeerg, 94), Y, S.., Paye, W. V. & Dmask, P. A. 011). Resdetal heat pump heatg perfrmace wth sgle faults mpsed. Appled hermal Egeerg, 315), Zha, Y., We, J. & Wag, S. 015). Dagstc Bayesa etwrks fr dagsg ar hadlg uts faults Part II: Faults cls ad sesrs. Appled hermal Egeerg, 90, ACKNOWLEDGEMEN he research wrk preseted ths paper s supprted by Natal Natural Scece Fudat f Cha Prect ), ad s fuded by Wuha Scece ad echlgy Bureau f Cha Prect ), ad s fuded by Beg Key Lab f eatg, Gas Supply, Vetlatg ad Ar Cdtg Egeerg Prect NR016K0), Cha, ad s supprted by state key labratry f cmpressr techlgy, Cha. 4 th Iteratal gh Perfrmace Buldgs Cferece at Purdue, July 11-14, 016
Data Mining: Concepts and Techniques
Data Mg: cepts ad Techques 3 rd ed. hapter 10 1 Evaluat f lusterg lusterg evaluat assesses the feasblty f clusterg aalyss a data set ad the qualty f the results geerated by a clusterg methd. Three mar
More informationExergy Analysis of Large ME-TVC Desalination System
Exergy Aalyss f arge ME-V esalat System Awar O. Bamer Water & Eergy Prgram\Research rectrate Kuwat udat fr the Advacemet f Sceces (KAS) he 0 th Gulf Water ferece, -4 Aprl 0, ha- Qatar Outles Itrduct Prcess
More informationBasics of heteroskedasticity
Sect 8 Heterskedastcty ascs f heterskedastcty We have assumed up t w ( ur SR ad MR assumpts) that the varace f the errr term was cstat acrss bservats Ths s urealstc may r mst ecmetrc applcats, especally
More informationThe Simple Linear Regression Model: Theory
Chapter 3 The mple Lear Regress Mdel: Ther 3. The mdel 3.. The data bservats respse varable eplaatr varable : : Plttg the data.. Fgure 3.: Dsplag the cable data csdered b Che at al (993). There are 79
More informationThe fuzzy decision of transformer economic operation
The fuzzy decs f trasfrmer ecmc perat WENJUN ZHNG, HOZHONG CHENG, HUGNG XIONG, DEXING JI Departmet f Electrcal Egeerg hagha Jatg Uversty 954 Huasha Rad, 3 hagha P. R. CHIN bstract: - Ths paper presets
More informationCHAPTER 5 ENTROPY GENERATION Instructor: Prof. Dr. Uğur Atikol
CAPER 5 ENROPY GENERAION Istructr: Pr. Dr. Uğur Atkl Chapter 5 Etrpy Geerat (Exergy Destruct Outle st Avalable rk Cycles eat ege cycles Rergerat cycles eat pump cycles Nlw Prcesses teady-flw Prcesses Exergy
More informationPY3101 Optics. Learning objectives. Wave propagation in anisotropic media Poynting walk-off The index ellipsoid Birefringence. The Index Ellipsoid
The Ide Ellpsd M.P. Vaugha Learg bjectves Wave prpagat astrpc meda Ptg walk-ff The de ellpsd Brefrgece 1 Wave prpagat astrpc meda The wave equat Relatve permttvt I E. Assumg free charges r currets E. Substtutg
More informationGoal of the Lecture. Lecture Structure. FWF 410: Analysis of Habitat Data I: Definitions and Descriptive Statistics
FWF : Aalyss f Habtat Data I: Defts ad Descrptve tatstcs Number f Cveys A A B Bur Dsk Bur/Dsk Habtat Treatmet Matthew J. Gray, Ph.D. Cllege f Agrcultural ceces ad Natural Resurces Uversty f Teessee-Kvlle
More informationEstimation of Stress- Strength Reliability model using finite mixture of exponential distributions
Iteratoal Joural of Computatoal Egeerg Research Vol, 0 Issue, Estmato of Stress- Stregth Relablty model usg fte mxture of expoetal dstrbutos K.Sadhya, T.S.Umamaheswar Departmet of Mathematcs, Lal Bhadur
More informationSimulation Output Analysis
Smulato Output Aalyss Summary Examples Parameter Estmato Sample Mea ad Varace Pot ad Iterval Estmato ermatg ad o-ermatg Smulato Mea Square Errors Example: Sgle Server Queueg System x(t) S 4 S 4 S 3 S 5
More informationIntroduction to local (nonparametric) density estimation. methods
Itroducto to local (oparametrc) desty estmato methods A slecture by Yu Lu for ECE 66 Sprg 014 1. Itroducto Ths slecture troduces two local desty estmato methods whch are Parze desty estmato ad k-earest
More informationLoad Frequency Control in Interconnected Power System Using Modified Dynamic Neural Networks
Prceedgs f the 5th Medterraea Cferece Ctrl & Autmat, July 7-9, 007, Athes - Greece 6-0 Lad Frequecy Ctrl tercected Pwer System Usg Mdfed Dyamc Neural Netwrks K.Sabah, M.A.Neku, M.eshehlab, M.Alyar ad M.Masur
More informationSummary of the lecture in Biostatistics
Summary of the lecture Bostatstcs Probablty Desty Fucto For a cotuos radom varable, a probablty desty fucto s a fucto such that: 0 dx a b) b a dx A probablty desty fucto provdes a smple descrpto of the
More informationbest estimate (mean) for X uncertainty or error in the measurement (systematic, random or statistical) best
Error Aalyss Preamble Wheever a measuremet s made, the result followg from that measuremet s always subject to ucertaty The ucertaty ca be reduced by makg several measuremets of the same quatty or by mprovg
More informationTo use adaptive cluster sampling we must first make some definitions of the sampling universe:
8.3 ADAPTIVE SAMPLING Most of the methods dscussed samplg theory are lmted to samplg desgs hch the selecto of the samples ca be doe before the survey, so that oe of the decsos about samplg deped ay ay
More informationComparing Different Estimators of three Parameters for Transmuted Weibull Distribution
Global Joural of Pure ad Appled Mathematcs. ISSN 0973-768 Volume 3, Number 9 (207), pp. 55-528 Research Ida Publcatos http://www.rpublcato.com Comparg Dfferet Estmators of three Parameters for Trasmuted
More informationCh5 Appendix Q-factor and Smith Chart Matching
h5 Appedx -factr ad mth hart Matchg 5B-1 We-ha a udwg, F rcut Desg Thery ad Applcat, hapter 8 Frequecy espse f -type Matchg Netwrks 5B- Fg.8-8 Tw desg realzats f a -type matchg etwrk.65pf, 80 f 1 GHz Fg.8-9
More informationEconometric Methods. Review of Estimation
Ecoometrc Methods Revew of Estmato Estmatg the populato mea Radom samplg Pot ad terval estmators Lear estmators Ubased estmators Lear Ubased Estmators (LUEs) Effcecy (mmum varace) ad Best Lear Ubased Estmators
More informationStatistics MINITAB - Lab 5
Statstcs 10010 MINITAB - Lab 5 PART I: The Correlato Coeffcet Qute ofte statstcs we are preseted wth data that suggests that a lear relatoshp exsts betwee two varables. For example the plot below s of
More informationNonlinear identification of synchronous generators using a local model approach
Seyyed Salma AHMADI, Mehd KARRARI Amrkabr Uversty f echlgy lear detfcat f sychrus geeratrs usg a lcal mdel apprach Abstract. A ew teratve apprach s prpsed t mdel sychrus geeratrs. Dfferet lcal structures
More informationMidterm Exam 1, section 1 (Solution) Thursday, February hour, 15 minutes
coometrcs, CON Sa Fracsco State Uversty Mchael Bar Sprg 5 Mdterm am, secto Soluto Thursday, February 6 hour, 5 mutes Name: Istructos. Ths s closed book, closed otes eam.. No calculators of ay kd are allowed..
More informationAnalysis of Variance with Weibull Data
Aalyss of Varace wth Webull Data Lahaa Watthaacheewaul Abstract I statstcal data aalyss by aalyss of varace, the usual basc assumptos are that the model s addtve ad the errors are radomly, depedetly, ad
More informationWu-Hausman Test: But if X and ε are independent, βˆ. ECON 324 Page 1
Wu-Hausma Test: Detectg Falure of E( ε X ) Caot drectly test ths assumpto because lack ubased estmator of ε ad the OLS resduals wll be orthogoal to X, by costructo as ca be see from the momet codto X'
More informationUNIVERSITY OF OSLO DEPARTMENT OF ECONOMICS
UNIVERSITY OF OSLO DEPARTMENT OF ECONOMICS Postpoed exam: ECON430 Statstcs Date of exam: Jauary 0, 0 Tme for exam: 09:00 a.m. :00 oo The problem set covers 5 pages Resources allowed: All wrtte ad prted
More informationMultivariate Transformation of Variables and Maximum Likelihood Estimation
Marquette Uversty Multvarate Trasformato of Varables ad Maxmum Lkelhood Estmato Dael B. Rowe, Ph.D. Assocate Professor Departmet of Mathematcs, Statstcs, ad Computer Scece Copyrght 03 by Marquette Uversty
More informationON THE LAGRANGIAN RHEONOMIC MECHANICAL SYSTEMS
THE PUBISHING HOUSE PROCEEDINGS OF THE ROMANIAN ACADEMY, Seres A, OF THE ROMANIAN ACADEMY Vlume 1, Number 1/9, pp. ON THE AGRANGIAN RHEONOMIC MECHANICA SYSTEMS Radu MIRON*, Tmak KAWAGUCHI**, Hrak KAWAGUCHI***
More informationDescriptive Statistics
Page Techcal Math II Descrptve Statstcs Descrptve Statstcs Descrptve statstcs s the body of methods used to represet ad summarze sets of data. A descrpto of how a set of measuremets (for eample, people
More informationMultiple Linear Regression Analysis
LINEA EGESSION ANALYSIS MODULE III Lecture - 4 Multple Lear egresso Aalyss Dr. Shalabh Departmet of Mathematcs ad Statstcs Ida Isttute of Techology Kapur Cofdece terval estmato The cofdece tervals multple
More informationA Method for Damping Estimation Based On Least Square Fit
Amerca Joural of Egeerg Research (AJER) 5 Amerca Joural of Egeerg Research (AJER) e-issn: 3-847 p-issn : 3-936 Volume-4, Issue-7, pp-5-9 www.ajer.org Research Paper Ope Access A Method for Dampg Estmato
More informationb. There appears to be a positive relationship between X and Y; that is, as X increases, so does Y.
.46. a. The frst varable (X) s the frst umber the par ad s plotted o the horzotal axs, whle the secod varable (Y) s the secod umber the par ad s plotted o the vertcal axs. The scatterplot s show the fgure
More informationAbstract. Introduction
THE IMPACT OF USING LOG-ERROR CERS OUTSIDE THE DATA RANGE AND PING FACTOR By Dr. Shu-Pg Hu Teclte Research Ic. 566 Hllster Ave. Ste. 30 Sata Barbara, CA 93 Abstract Ths paper dscusses the prs ad cs f usg
More informationCHAPTER VI Statistical Analysis of Experimental Data
Chapter VI Statstcal Aalyss of Expermetal Data CHAPTER VI Statstcal Aalyss of Expermetal Data Measuremets do ot lead to a uque value. Ths s a result of the multtude of errors (maly radom errors) that ca
More informationChapter 5 Properties of a Random Sample
Lecture 6 o BST 63: Statstcal Theory I Ku Zhag, /0/008 Revew for the prevous lecture Cocepts: t-dstrbuto, F-dstrbuto Theorems: Dstrbutos of sample mea ad sample varace, relatoshp betwee sample mea ad sample
More informationTHE ROYAL STATISTICAL SOCIETY HIGHER CERTIFICATE
THE ROYAL STATISTICAL SOCIETY 00 EXAMINATIONS SOLUTIONS HIGHER CERTIFICATE PAPER I STATISTICAL THEORY The Socety provdes these solutos to assst caddates preparg for the examatos future years ad for the
More informationMidterm Exam 1, section 2 (Solution) Thursday, February hour, 15 minutes
coometrcs, CON Sa Fracsco State Uverst Mchael Bar Sprg 5 Mdterm xam, secto Soluto Thursda, Februar 6 hour, 5 mutes Name: Istructos. Ths s closed book, closed otes exam.. No calculators of a kd are allowed..
More informationLecture 3. Sampling, sampling distributions, and parameter estimation
Lecture 3 Samplg, samplg dstrbutos, ad parameter estmato Samplg Defto Populato s defed as the collecto of all the possble observatos of terest. The collecto of observatos we take from the populato s called
More informationChapter 14 Logistic Regression Models
Chapter 4 Logstc Regresso Models I the lear regresso model X β + ε, there are two types of varables explaatory varables X, X,, X k ad study varable y These varables ca be measured o a cotuous scale as
More informationByeong-Joo Lee
yeg-j Lee OSECH - MSE calphad@pstech.ac.r yeg-j Lee www.pstech.ac.r/~calphad Fudametals Mcrscpc vs. Macrscpc Vew t State fuct vs. rcess varable Frst Law f hermdyamcs Specal prcesses 1. Cstat-Vlume rcess:
More informationCh5 Appendix Q-factor and Smith Chart Matching
Ch5 Appedx -factr ad mth Chart Matchg 5B-1 We-Cha a udwg, F Crcut Deg hery ad Applcat, Chapter 8 -type matchg etwrk w-cmpet Matchg Netwrk hee etwrk ue tw reactve cmpet t trafrm the lad mpedace t the dered
More informationSimple Linear Regression
Statstcal Methods I (EST 75) Page 139 Smple Lear Regresso Smple regresso applcatos are used to ft a model descrbg a lear relatoshp betwee two varables. The aspects of least squares regresso ad correlato
More informationENGI 3423 Simple Linear Regression Page 12-01
ENGI 343 mple Lear Regresso Page - mple Lear Regresso ometmes a expermet s set up where the expermeter has cotrol over the values of oe or more varables X ad measures the resultg values of aother varable
More informationECONOMETRIC THEORY. MODULE VIII Lecture - 26 Heteroskedasticity
ECONOMETRIC THEORY MODULE VIII Lecture - 6 Heteroskedastcty Dr. Shalabh Departmet of Mathematcs ad Statstcs Ida Isttute of Techology Kapur . Breusch Paga test Ths test ca be appled whe the replcated data
More informationChapter 2 - Free Vibration of Multi-Degree-of-Freedom Systems - II
CEE49b Chapter - Free Vbrato of Mult-Degree-of-Freedom Systems - II We ca obta a approxmate soluto to the fudametal atural frequecy through a approxmate formula developed usg eergy prcples by Lord Raylegh
More informationChapter 11 The Analysis of Variance
Chapter The Aalyss of Varace. Oe Factor Aalyss of Varace. Radomzed Bloc Desgs (ot for ths course) NIPRL . Oe Factor Aalyss of Varace.. Oe Factor Layouts (/4) Suppose that a expermeter s terested populatos
More informationd dt d d dt dt Also recall that by Taylor series, / 2 (enables use of sin instead of cos-see p.27 of A&F) dsin
Learzato of the Swg Equato We wll cover sectos.5.-.6 ad begg of Secto 3.3 these otes. 1. Sgle mache-fte bus case Cosder a sgle mache coected to a fte bus, as show Fg. 1 below. E y1 V=1./_ Fg. 1 The admttace
More informationLecture 2. Basic Semiconductor Physics
Lecture Basc Semcductr Physcs I ths lecture yu wll lear: What are semcductrs? Basc crystal structure f semcductrs Electrs ad hles semcductrs Itrsc semcductrs Extrsc semcductrs -ded ad -ded semcductrs Semcductrs
More informationInvestigation of Partially Conditional RP Model with Response Error. Ed Stanek
Partally Codtoal Radom Permutato Model 7- vestgato of Partally Codtoal RP Model wth Respose Error TRODUCTO Ed Staek We explore the predctor that wll result a smple radom sample wth respose error whe a
More informationANALYSIS OF HYBRID VENTILATION PERFORMANCE IN FRANCE
ANALYSIS OF HYBRID VENTILATION PERFORMANCE IN FRANCE Flrece Cr ad Chrsta Iard LEPTAB, Uversté de Pôle Sceces et Techlge, Aveue Mchel Crépeau F-742 Cedex, Frace Emals: fcr@uv-lr.fr ad card@uv-lr.fr Eghth
More informationSolving Constrained Flow-Shop Scheduling. Problems with Three Machines
It J Cotemp Math Sceces, Vol 5, 2010, o 19, 921-929 Solvg Costraed Flow-Shop Schedulg Problems wth Three Maches P Pada ad P Rajedra Departmet of Mathematcs, School of Advaced Sceces, VIT Uversty, Vellore-632
More informationLinear Regression with One Regressor
Lear Regresso wth Oe Regressor AIM QA.7. Expla how regresso aalyss ecoometrcs measures the relatoshp betwee depedet ad depedet varables. A regresso aalyss has the goal of measurg how chages oe varable,
More informationProcessing of Information with Uncertain Boundaries Fuzzy Sets and Vague Sets
Processg of Iformato wth Ucerta odares Fzzy Sets ad Vage Sets JIUCHENG XU JUNYI SHEN School of Electroc ad Iformato Egeerg X'a Jaotog Uversty X'a 70049 PRCHIN bstract: - I the paper we aalyze the relatoshps
More informationArithmetic Mean and Geometric Mean
Acta Mathematca Ntresa Vol, No, p 43 48 ISSN 453-6083 Arthmetc Mea ad Geometrc Mea Mare Varga a * Peter Mchalča b a Departmet of Mathematcs, Faculty of Natural Sceces, Costate the Phlosopher Uversty Ntra,
More informationApplication of Matrix Iteration for Determining the Fundamental Frequency of Vibration of a Continuous Beam
Iteratal Jural f Egeerg Research ad Develpet e-issn: 78-67, p-issn : 78-8, www.jerd.c Vlue 4, Issue (Nveber ), PP. -6 Applcat f Matrx Iterat fr Deterg the Fudaetal Frequecy f Vbrat f a Ctuus Bea S. Sule,
More informationMEASURES OF DISPERSION
MEASURES OF DISPERSION Measure of Cetral Tedecy: Measures of Cetral Tedecy ad Dsperso ) Mathematcal Average: a) Arthmetc mea (A.M.) b) Geometrc mea (G.M.) c) Harmoc mea (H.M.) ) Averages of Posto: a) Meda
More informationSampling Theory MODULE X LECTURE - 35 TWO STAGE SAMPLING (SUB SAMPLING)
Samplg Theory ODULE X LECTURE - 35 TWO STAGE SAPLIG (SUB SAPLIG) DR SHALABH DEPARTET OF ATHEATICS AD STATISTICS IDIA ISTITUTE OF TECHOLOG KAPUR Two stage samplg wth uequal frst stage uts: Cosder two stage
More informationESS Line Fitting
ESS 5 014 17. Le Fttg A very commo problem data aalyss s lookg for relatoshpetwee dfferet parameters ad fttg les or surfaces to data. The smplest example s fttg a straght le ad we wll dscuss that here
More information{ }{ ( )} (, ) = ( ) ( ) ( ) Chapter 14 Exercises in Sampling Theory. Exercise 1 (Simple random sampling): Solution:
Chapter 4 Exercses Samplg Theory Exercse (Smple radom samplg: Let there be two correlated radom varables X ad A sample of sze s draw from a populato by smple radom samplg wthout replacemet The observed
More informationAn Introduction to. Support Vector Machine
A Itroducto to Support Vector Mache Support Vector Mache (SVM) A classfer derved from statstcal learg theory by Vapk, et al. 99 SVM became famous whe, usg mages as put, t gave accuracy comparable to eural-etwork
More informationOrdinary Least Squares Regression. Simple Regression. Algebra and Assumptions.
Ordary Least Squares egresso. Smple egresso. Algebra ad Assumptos. I ths part of the course we are gog to study a techque for aalysg the lear relatoshp betwee two varables Y ad X. We have pars of observatos
More informationMaximum Likelihood Estimation
Marquette Uverst Maxmum Lkelhood Estmato Dael B. Rowe, Ph.D. Professor Departmet of Mathematcs, Statstcs, ad Computer Scece Coprght 08 b Marquette Uverst Maxmum Lkelhood Estmato We have bee sag that ~
More informationComparison of Dual to Ratio-Cum-Product Estimators of Population Mean
Research Joural of Mathematcal ad Statstcal Sceces ISS 30 6047 Vol. 1(), 5-1, ovember (013) Res. J. Mathematcal ad Statstcal Sc. Comparso of Dual to Rato-Cum-Product Estmators of Populato Mea Abstract
More informationEVALUATION OF FUNCTIONAL INTEGRALS BY MEANS OF A SERIES AND THE METHOD OF BOREL TRANSFORM
EVALUATION OF FUNCTIONAL INTEGRALS BY MEANS OF A SERIES AND THE METHOD OF BOREL TRANSFORM Jose Javer Garca Moreta Ph. D research studet at the UPV/EHU (Uversty of Basque coutry) Departmet of Theoretcal
More informationAnalysis of a Positive Output Super-Lift Luo Boost Converter
Ausha eade et al. t. Jural f Egeerg esearch ad Applicats SSN: 8-96, l. 6, ssue, (Part - 5) February 06, pp.7-78 ESEACH ACE www.ijera.cm OPEN ACCESS Aalys f a Psitive Output Super-ift u Bst Cverter Ausha
More informationLecture Notes Types of economic variables
Lecture Notes 3 1. Types of ecoomc varables () Cotuous varable takes o a cotuum the sample space, such as all pots o a le or all real umbers Example: GDP, Polluto cocetrato, etc. () Dscrete varables fte
More informationKernel-based Methods and Support Vector Machines
Kerel-based Methods ad Support Vector Maches Larr Holder CptS 570 Mache Learg School of Electrcal Egeerg ad Computer Scece Washgto State Uverst Refereces Muller et al. A Itroducto to Kerel-Based Learg
More informationLINEAR REGRESSION ANALYSIS
LINEAR REGRESSION ANALYSIS MODULE V Lecture - Correctg Model Iadequaces Through Trasformato ad Weghtg Dr. Shalabh Departmet of Mathematcs ad Statstcs Ida Isttute of Techology Kapur Aalytcal methods for
More informationThird handout: On the Gini Index
Thrd hadout: O the dex Corrado, a tala statstca, proposed (, 9, 96) to measure absolute equalt va the mea dfferece whch s defed as ( / ) where refers to the total umber of dvduals socet. Assume that. The
More informationKernels. Nuno Vasconcelos ECE Department, UCSD
Kerels Nu Vasels ECE Departmet UCSD Prpal mpet aalyss Dmesalty reut: Last tme we saw that whe the ata lves a subspae t s best t esg ur learg algrthms ths subspae D subspae 3D y φ φ λ λ y ths a be e by
More informationLecture 3 Probability review (cont d)
STATS 00: Itroducto to Statstcal Iferece Autum 06 Lecture 3 Probablty revew (cot d) 3. Jot dstrbutos If radom varables X,..., X k are depedet, the ther dstrbuto may be specfed by specfyg the dvdual dstrbuto
More informationVOL. 3, NO. 11, November 2013 ISSN ARPN Journal of Science and Technology All rights reserved.
VOL., NO., November 0 ISSN 5-77 ARPN Joural of Scece ad Techology 0-0. All rghts reserved. http://www.ejouralofscece.org Usg Square-Root Iverted Gamma Dstrbuto as Pror to Draw Iferece o the Raylegh Dstrbuto
More informationAnalyzing Fuzzy System Reliability Using Vague Set Theory
Iteratoal Joural of Appled Scece ad Egeerg 2003., : 82-88 Aalyzg Fuzzy System Relablty sg Vague Set Theory Shy-Mg Che Departmet of Computer Scece ad Iformato Egeerg, Natoal Tawa versty of Scece ad Techology,
More informationMATHEMATICAL PROGRAMMING-BASED PERTURBATION ANALYSIS FOR GI/G/1 QUEUES. He Zhang Wai Kin (Victor) Chan
Prceedgs f the 007 Wter Smulat Cferece S. G. Heders,. ller, M.-H. Hseh, J. Shrtle, J. D. ew, ad R. R. art, eds. MAHEMAICAL PROGRAMMING-ASED PERURAION ANALYSIS FOR GI/G/ QUEUES He Zhag Wa K (Vctr Cha Departmet
More informationε. Therefore, the estimate
Suggested Aswers, Problem Set 3 ECON 333 Da Hugerma. Ths s ot a very good dea. We kow from the secod FOC problem b) that ( ) SSE / = y x x = ( ) Whch ca be reduced to read y x x = ε x = ( ) The OLS model
More informationFunctions of Random Variables
Fuctos of Radom Varables Chapter Fve Fuctos of Radom Varables 5. Itroducto A geeral egeerg aalyss model s show Fg. 5.. The model output (respose) cotas the performaces of a system or product, such as weght,
More informationSound Absorption Characteristics of Membrane- Based Sound Absorbers
Purdue e-pubs Publicatis f the Ray W. Schl f Mechaical Egieerig 8-28-2003 Sud Absrpti Characteristics f Membrae- Based Sud Absrbers J Stuart Blt, blt@purdue.edu Jih Sg Fllw this ad additial wrks at: http://dcs.lib.purdue.edu/herrick
More informationoutput units hidden units input units
Part-f-Speech Taggg wth Neural Netwrks Helmut Schmd y Isttut fur maschelle Sprachverarbetug, Uverstat Stuttgart, Azebergstr.12, 70174 Stuttgart, Germay, schmd@ms.u-stuttgart.de Tpc area: large text crpra,
More informationEvaluation of uncertainty in measurements
Evaluato of ucertaty measuremets Laboratory of Physcs I Faculty of Physcs Warsaw Uversty of Techology Warszawa, 05 Itroducto The am of the measuremet s to determe the measured value. Thus, the measuremet
More informationA New Family of Transformations for Lifetime Data
Proceedgs of the World Cogress o Egeerg 4 Vol I, WCE 4, July - 4, 4, Lodo, U.K. A New Famly of Trasformatos for Lfetme Data Lakhaa Watthaacheewakul Abstract A famly of trasformatos s the oe of several
More informationBayes Interval Estimation for binomial proportion and difference of two binomial proportions with Simulation Study
IJIEST Iteratoal Joural of Iovatve Scece, Egeerg & Techology, Vol. Issue 5, July 04. Bayes Iterval Estmato for bomal proporto ad dfferece of two bomal proportos wth Smulato Study Masoud Gaj, Solmaz hlmad
More informationLecture 8: Linear Regression
Lecture 8: Lear egresso May 4, GENOME 56, Sprg Goals Develop basc cocepts of lear regresso from a probablstc framework Estmatg parameters ad hypothess testg wth lear models Lear regresso Su I Lee, CSE
More informationStatistics Descriptive and Inferential Statistics. Instructor: Daisuke Nagakura
Statstcs Descrptve ad Iferetal Statstcs Istructor: Dasuke Nagakura (agakura@z7.keo.jp) 1 Today s topc Today, I talk about two categores of statstcal aalyses, descrptve statstcs ad feretal statstcs, ad
More informationLecture 7. Confidence Intervals and Hypothesis Tests in the Simple CLR Model
Lecture 7. Cofdece Itervals ad Hypothess Tests the Smple CLR Model I lecture 6 we troduced the Classcal Lear Regresso (CLR) model that s the radom expermet of whch the data Y,,, K, are the outcomes. The
More informationAnalysis of System Performance IN2072 Chapter 5 Analysis of Non Markov Systems
Char for Network Archtectures ad Servces Prof. Carle Departmet of Computer Scece U Müche Aalyss of System Performace IN2072 Chapter 5 Aalyss of No Markov Systems Dr. Alexader Kle Prof. Dr.-Ig. Georg Carle
More informationPrediction By Regression and kriging For Spatial Data With Application
Predct By Regress ad krgg r Spatal Data Wt Applcat امللخص Abstract s stdy deals wt te predct f te -statary spatal stcastc prcess. e predct s de by tw tecqes wc are regress tecqe geeralzed least sqare estmat
More informationComparison of Parameters of Lognormal Distribution Based On the Classical and Posterior Estimates
Joural of Moder Appled Statstcal Methods Volume Issue Artcle 8 --03 Comparso of Parameters of Logormal Dstrbuto Based O the Classcal ad Posteror Estmates Raja Sulta Uversty of Kashmr, Sragar, Ida, hamzasulta8@yahoo.com
More informationA New Method for Solving Integer Linear. Programming Problems with Fuzzy Variables
Appled Mathematcal Scences, Vl. 4, 00, n. 0, 997-004 A New Methd fr Slvng Integer Lnear Prgrammng Prblems wth Fuzzy Varables P. Pandan and M. Jayalakshm Department f Mathematcs, Schl f Advanced Scences,
More informationPermutation Tests for More Than Two Samples
Permutato Tests for ore Tha Two Samples Ferry Butar Butar, Ph.D. Abstract A F statstc s a classcal test for the aalyss of varace where the uderlyg dstrbuto s a ormal. For uspecfed dstrbutos, the permutato
More informationFault Diagnosis Using Feature Vectors and Fuzzy Fault Pattern Rulebase
Fault Dagoss Usg Feature Vectors ad Fuzzy Fault Patter Rulebase Prepared by: FL Lews Updated: Wedesday, ovember 03, 004 Feature Vectors The requred puts for the dagostc models are termed the feature vectors
More informationChapter 8. Inferences about More Than Two Population Central Values
Chapter 8. Ifereces about More Tha Two Populato Cetral Values Case tudy: Effect of Tmg of the Treatmet of Port-We tas wth Lasers ) To vestgate whether treatmet at a youg age would yeld better results tha
More informationX X X E[ ] E X E X. is the ()m n where the ( i,)th. j element is the mean of the ( i,)th., then
Secto 5 Vectors of Radom Varables Whe workg wth several radom varables,,..., to arrage them vector form x, t s ofte coveet We ca the make use of matrx algebra to help us orgaze ad mapulate large umbers
More informationStatistical modelling and latent variables (2)
Statstcal modellg ad latet varables (2 Mxg latet varables ad parameters statstcal erece Trod Reta (Dvso o statstcs ad surace mathematcs, Departmet o Mathematcs, Uversty o Oslo State spaces We typcally
More informationSPECIAL CONSIDERATIONS FOR VOLUMETRIC Z-TEST FOR PROPORTIONS
SPECIAL CONSIDERAIONS FOR VOLUMERIC Z-ES FOR PROPORIONS Oe s stctve reacto to the questo of whether two percetages are sgfcatly dfferet from each other s to treat them as f they were proportos whch the
More informationn -dimensional vectors follow naturally from the one
B. Vectors ad sets B. Vectors Ecoomsts study ecoomc pheomea by buldg hghly stylzed models. Uderstadg ad makg use of almost all such models requres a hgh comfort level wth some key mathematcal sklls. I
More informationf f... f 1 n n (ii) Median : It is the value of the middle-most observation(s).
CHAPTER STATISTICS Pots to Remember :. Facts or fgures, collected wth a defte pupose, are called Data.. Statstcs s the area of study dealg wth the collecto, presetato, aalyss ad terpretato of data.. The
More information= 1. UCLA STAT 13 Introduction to Statistical Methods for the Life and Health Sciences. Parameters and Statistics. Measures of Centrality
UCLA STAT Itroducto to Statstcal Methods for the Lfe ad Health Sceces Istructor: Ivo Dov, Asst. Prof. of Statstcs ad Neurology Teachg Assstats: Fred Phoa, Krste Johso, Mg Zheg & Matlda Hseh Uversty of
More informationto the estimation of total sensitivity indices
Applcato of the cotrol o varate ate techque to the estmato of total sestvty dces S KUCHERENKO B DELPUECH Imperal College Lodo (UK) skuchereko@mperalacuk B IOOSS Electrcté de Frace (Frace) S TARANTOLA Jot
More informationClass 13,14 June 17, 19, 2015
Class 3,4 Jue 7, 9, 05 Pla for Class3,4:. Samplg dstrbuto of sample mea. The Cetral Lmt Theorem (CLT). Cofdece terval for ukow mea.. Samplg Dstrbuto for Sample mea. Methods used are based o CLT ( Cetral
More informationBAL-001-AB-0a Real Power Balancing Control Performance
Alberta Relablty Stadards Resource ad Demad Balacg BAL-00-AB-0a. Purpose BAL-00-AB-0a Real Power Balacg Cotrol Performace The purpose of ths relablty stadard s to mata WECC steady-state frequecy wth defed
More informationBAYESIAN INFERENCES FOR TWO PARAMETER WEIBULL DISTRIBUTION
Iteratoal Joural of Mathematcs ad Statstcs Studes Vol.4, No.3, pp.5-39, Jue 06 Publshed by Europea Cetre for Research Trag ad Developmet UK (www.eajourals.org BAYESIAN INFERENCES FOR TWO PARAMETER WEIBULL
More informationTHE ROYAL STATISTICAL SOCIETY GRADUATE DIPLOMA
THE ROYAL STATISTICAL SOCIETY EXAMINATIONS SOLUTIONS GRADUATE DIPLOMA PAPER II STATISTICAL THEORY & METHODS The Socety provdes these solutos to assst caddates preparg for the examatos future years ad for
More information