A Novel Approach to Model Generation for Heterogeneous Data Classification
|
|
- Natalie Stevenson
- 6 years ago
- Views:
Transcription
1 A Novel Approach o Model Generaon for Heerogeneous Daa Classfcaon Rong Jn*, Huan Lu *Dep. of Compuer Scence and Engneerng, Mchgan Sae Unversy, Eas Lansng, MI rongn@cse.msu.edu Deparmen of Compuer Scence and Engneerng, Arzona Sae Unversy, Tempe, AZ hlu@asu.edu Absrac Ensemble mehods such as baggng and boosng have been successfully appled o classfcaon problems. Two mporan ssues assocaed wh an ensemble approach are: how o generae models o consruc an ensemble, and how o combne hem for classfcaon. In hs paper, we focus on he problem of model generaon for heerogeneous daa classfcaon. If we could paron heerogeneous daa no a number of homogeneous parons, we wll lkely generae relable and accurae classfcaon models over he homogeneous parons. We examne dfferen ways of formng homogeneous subses and propose a novel mehod ha allows a daa pon o be assgned mulple mes n order o generae homogeneous parons for ensemble learnng. We presen he deals of he new algorhm and emprcal sudes over he UCI benchmark daases and daases of mage classfcaon, and show ha he proposed approach s effecve for heerogeneous daa classfcaon. 1 Inroducon Ensemble approaches such as baggng and boosng have been successfully appled o many classfcaon problems [Deerch, 2000; Bauer and ohav, 1999]. The basc dea of ensemble mehods s o consruc a number of classfers over ranng daa and hen classfy new daa pons by akng a (weghed) voe of her predcons. Thus, wo mporan ssues assocaed wh an ensemble approach are: 1) how o generae accurae ye dverse classfcaon models, and 2) how o combne he models for ensemble classfcaon. Dverse classfers ensure good ensembles [Qunlan, 1996]. In hs paper, we focus on he frs ssue wh an emphass on heerogeneous daa classfcaon. Heerogeneous daa classfcaon refers o he problem when npu daa of a sngle class are wdely dsrbued no mulple modes. I arses when ranng daa are colleced under dfferen envronmens or hrough dfferen sources. An example of heerogeneous daa classfcaon s mage classfcaon, n whch labeled mages are acqured from mulple resources and exhb dsparae characerscs. For nsance, some mages are black and whe, and ohers are colorful. A wdely used approach for consrucng an ensemble of models s o sample dfferen subses from he ranng daa and creae a classfcaon model for each subse. Baggng [Bremann, 1996] and AdaBoos [Schapre and Snger, 1999] are wo represenave mehods n hs caegory. Baggng randomly draws samples from he ranng daa wh replacemen and AdaBoos samples ranng daa accordng o a dynamcally changed dsrbuon, whch s updaed by pung more wegh on he msclassfed examples and smaller weghs on he correcly classfed examples. Clearly, boh mehods do no rea homogeneous daa and heerogeneous daa dfferenly. For ensemble mehods o work effecvely on heerogeneous daa, one nuve soluon s o frs dvde he heerogeneous daa no a se of homogeneous parons and hen o creae a model for each paron of daa. Member classfers bul wh dfferen homogeneous parons wll lkely resul n good dversy of an ensemble. One way o realze hs homogeney-based paron s o employ sandard cluserng algorhms, such as -means [Hargan and Wong, 1979] and he EM cluserng algorhm [Celeux and Govaer, 1992]. An example s he Gaussan Mxure Model (GMM). Bu, n general, here are wo problems wh hs smple cluserng approach: Sngle cluser membershp. Mos cluserng algorhms assume ha cluser membershp s muually exclusve and each daa pon can only belong o a sngle cluser. Even hough he EM cluserng algorhm allows sof membershp for a daa pon, n he resulng clusers, each daa pon sll only belongs o a sngle cluser [Wen and Frank, 2000]. Therefore, when we use hese cluserng algorhms o paron daa, f he number of clusers s large and he subses of ranng daa formed by a cluserng algorhm are muually dson, some clusers may have a very small number of daa pons, whch can lead o unrelable classfcaon models. Ths s smlar o he daa fragmenaon problem occurred n decson ree nducon [Qunlan, 1993]. In conras, he subses of ranng daa produced by Baggng and AdaBoos are no muually dson. For example, n boosrap samplng, each subse conans around 63.2% of he orgnal ranng daa. Unbalanced cluser szes. Snce mos cluserng algorhms do no have conrol over cluser szes, unbalanced cluser szes resulng from cluserng canno be easly cor-
2 reced. When he resulng clusers have very dfferen szes, a classfer bul over a small cluser can be unrelable and hus degrade he performance of he ensemble n formng fnal ensemble classfcaon. On he conrary, boh Baggng and AdaBoos have daa samples of smlar szes when learnng dfferen models. Noe ha here have been prevous effors on balancng he szes of dfferen clusers, parcularly for specral cluserng algorhms (e.g., he normalzed cu algorhm [Mela & Sh, 2001]). Bu, snce he conrol of cluser sze comes ndrecly from he obecve funcon, he resulng clusers can sll have unbalanced szes. In sum, a cluserng approach may produce homogeneous daa parons, bu canno ensure smlar szes of dfferen parons; mehods lke Baggng can produce equally szed parons, bu parons are no homogeneous. Therefore, we need a novel approach o paronng daa no homogeneous subses of smlar szes n ensemble learnng for heerogeneous daa classfcaon. The goal of hs work s o dvde heerogeneous daa no homogeneous subses of smlar szes n order o generae relable and accurae classfcaon models. By focusng on homogeneous subses, we do no requre ha each daa pon belong o one subse; by ensurng smlar szes of daa subses, each classfcaon model can be bul wh a smlar number of daa pons. In hs paper, we propose a HISS (Homogeneous daa In Smlar Sze) algorhm specally desgned for he above purposes for heerogeneous daa classfcaon. Specfcally, HISS allows he user o specfy he sze of a subse. For example, he user can ask he algorhm o creae 20 subses wh each conanng 40% of he orgnal daa. Ths algorhm s smlar o he boosrap samplng procedure n ha boh he number of subses and he percenage of ranng daa covered by each cluser can be specfed and vared. However, dffers from he smple boosrap samplng procedure n ha pus he smlar daa pons no a sngle subse whle boosrap samplng randomly selecs daa o form a subse. Ths propery s mporan n ensemble learnng for classfyng heerogeneous daa. We wll use sraa for he homogeneous daa parons, and subses for daa parons resulng from random samplng. 2 Relaed Work There have been many prevous sudes on how o creae an ensemble of models. The mehods for consrucng an ensemble of models can be caegorzed no fve groups [Deerch, 2000]: 1) Bayesan mehods, whch creaes an ensemble of model by samplng hem from a esmaed poseror model dsrbuon; 2) Samplng ranng examples, whch creaes mulple subses of ranng examples and rans a classfer for each of he subses; 3) Samplng npu feaures, whch creaes a number of subses of he npu feaures and a classfer s bul for each subse of npu feaures; 4) Error correc oupu code (ECOC), whch conver a mulple class problem no a se of bnary class problems; 5) Inecng randomness, ha generaes ensembles of classfers by necng randomness no he learnng algorhm. Among he fve caegores, our work s closely relaed o he second one, whch creaes mulple classfers by samplng ranng examples. Imporan mehods n hs group nclude Baggng [Breman, 1996] and AdaBoos [Schapre and Snger, 1999]. Alhough hese mehods have been shown o be effecve for classfcaon, hey are no desgned o ake no accoun characerscs of heerogeneous daa. In hs paper, we propose HISS an algorhm ha consrucs homogeneous sraa from heerogeneous daa whle manans he nce propery of boosrap samplng procedure - each sraum conans a smlar number of daa pons. Anoher lne of research closely relaed o hs work s he sudy of cluserng algorhms. In general, cluserng algorhms can be caegorzed no paramerc approaches and non-paramerc approaches. The paramerc approach s o fnd a paramerc model ha mnmzes a cos funcon assocaed wh nsance-cluser assgnmens. Such mehods nclude he Mxure Model [Celeux and Govaer, 1992] and -means algorhm. For he non-paramerc approaches, a cos funcon s mnmzed by eher mergng wo separae clusers no a larger one or dvdng a cluser no wo smaller ones. The represenave examples of hs caegory are he agglomerave approach and he dvsve approach. Mos cluserng approaches assume ha each daa pon only belongs o a sngle cluser. Ths assumpon may no be approprae snce he ulmae goal of cluserng s o group smlar daa pons ogeher. When s unceran o assgn a daa pon o a sngle cluser, s beer off assgnng o mulple clusers. Alhough he radonal probablsc model and he fuzzy cluserng algorhm allow for mul- or sof-membershps, he uncerany of cluser membershp s only exploed durng he process of esmaon. In he resulng clusers, each daa pon s assgned o only a sngle cluser. Furhermore, mos cluserng algorhms do no have any conrol over he sze of clusers. Hence, he resulng clusers can be very unbalanced n sze and he clusers of oo small szes could be useless n learnng. 3 The HISS Algorhm for Model Generaon 3.1 From Probablsc Cluserng o HISS We frs descrbe he radonal probablsc cluserng algorhm, and hen nroduce algorhm HISS. The general dea of probablsc cluserng s o descrbe daa wh a mxure of generave models. Opmal parameers are usually obaned by maxmzng he lkelhood of daa usng he mxure model. Le n be he number of npu daa pons, be he number clusers, { x1, x2,..., x n } be he npu daa, and { m1, m2,..., m } be he underlyng models ha generae he daa. By assumng ha each daa pon s generaed from a mxure of models{ m1, m2,..., m }, we have he lkelhood of he daa wren as: n l({ m} 1, = ) = log p( x m) = 1 = 1 (1)
3 where px ( m ) s he lkelhood of generang x from he model m, and s he lkelhood for daa pon x o be n he -h cluser. Based on he assumpon ha each daa pon can only belong o a sngle cluser, we have consran = 1. An example of probablsc cluserng s he = 1 Gaussan Mxure Model (GMM), n whch boh px ( m ) are parameerzed as: = θ, and T 1 1 ( x µ ) Σ ( x µ ) px ( m) = exp d /2 1/ 2 ( 2 π ) Σ 2 where θ denoes he pror for he -h cluser, and and (2) µ and Σ are he mean and varance marx for he -h cluser, respecvely. Expecaon and Maxmzaon algorhm (EM) (Dempser e al, 1977) can be used o search for he opmal parameers. By removng he consran 1 = 1 =, we allow each daa pon o belong o mulple homogeneous clusers, or n shor, sraa. Hence, he opmzaon problem becomes n max l({ m} 1, ) log ( ) = = p x m m, = 1 = 1 subec o 0 1 for = 1,..., n, and, = 1,..., where all are consraned o beween 0 and 1 o manan he probably nerpreaon. I s easy o see ha he opmal soluon s o se all o be 1, whch means ha each daa pon s ncluded n every sraum. To avod he rval soluon for, we choose o enforce he percenage of ranng daa ha are covered by each cluser o be a predefned consanγ,.e., (3) 1 n, 0 1, for 1,..., 1 = γ γ = (4) = n Wh he above consran, we guaranee ha he number of daa pons ha suppor each sraum s around γ n. Compared o he sngle membershp consran, hs new consran has he followng wo advanages: 1) I does no assume ha each daa pon has o belong o one sraum. For hs new srafyng mehod, on average each daa pon can belong o γ number of sraa. Therefore, when γ s larger han one, each daa pon s allowed o be n more han one sraum smulaneously. 2) I ensures ha dfferen sraa have balanced numbers of daa pons. In conras o mos cluserng algorhms, he new algorhm ensures almos he same sze for each sraum. Ths s parcularly mporan o he research goal of hs paper - generang a relable and accurae ensemble for heerogeneous daa. By seng γ o be a reasonably large value (0.4 n hs work), we ensure ha each sraum has a suffcenly large number of examples for buldng a sascal learnng model. For laer reference, we refer hs new cluserng approach as HISS, whch sands for Homogeneous daa In Smlar Sze. 3.2 Opmzaon for HISS Pung Equaons (3) and (4) ogeher, we have: n max l({ m} 1, ) log ( ) = = p x m m, = 1 = 1 subec o 1 = γ, 0 γ 1, for = 1,..., n 0 1 for = 1,..., n, and, = 1,..., n = 1 Le us assume he Gaussan dsrbuon for px ( m ),.e., px ( m )~ Nµ (, σ ). Followng he dea of he EM algorhm, he dfference n he lkelhood of daa beween wo consecuve eraons s bound by: l({ m( + 1)} = 1, ( + 1)) l({ m( )} = 1, ( )) n ( + 1) = 1 = 1υ ()log + () (6) n px ( m( + 1)) = 1 = 1υ ()log px ( m( )) where υ s defned as (5) () p( x m()) υ () = k (7) () p( x m ()) = 1 k Thus, he opmal soluons for he mean and varance of Gaussan dsrbuon can be obaned as follows: n n 2 υ () () 1 x υ 2 1 x = = 2 n n υ () () 1 υ 1 = = µ ( + 1) =, σ ( + 1) = µ ( + 1) However, he opmal soluon for s raher dffcul o oban because of he nequaly consrans 0 1. Drecly opmzng he Equaon (6) wh only he equaly consran wll resul n he followng soluon for ( + 1) : ( 1) n υ = 1 υ () γ n + = (8) () Apparenly, he above soluon wll always be nonnegave f () s nonnegave. However, does no guaranee ha ( + 1) s no greaer han 1.
4 Fndng Opmal ( + 1) Inpus: υ () for = 1,..., n and = 1,..., Oupus: ( + 1) ha maxmzes Equaon (7). Inalzaon: ( + 1) = 0 for = 1,..., n and = 1,..., for each cluser do For all examples, se ( + 1) = 1 f ( + 1) > 1 Compue he probably mass s = γn { ( + 1) = 1} Re-compue υ () ( + 1) = s s.. ( + 1) < 1 υ () { ( + 1) < 1} whle ( s.. ( + 1) > 1 ) end Fgure 1: Algorhm for fndng opmal ( + 1) In order o sasfy he nequaly consrans 0 1, we use he T condons [Flecher, 1987] o effcenly adus he value of ( + 1). The basc dea s o rese o be 1 whenever he oupu from Equaon (10) volaes he consran 0 1. Afer he adusmen, we wll recompue ( + 1) ha are less han 1 usng Equaon (8). The procedure of adusng and recompung ( + 1) wll connue unl no ( + 1) volaes he consran. Fgure 1 shows he dealed seps for fndng he opmal soluon for ( + 1). Due o he space lm, he proof for he opmaly of he algorhm n Fgure 1 s no provded here. 3.3 Classfyng Heerogeneous Daa For classfcaon problems, heerogeneous daa can be found n many applcaons and n expermens: 1) Daa acqured from mulple sources. In many cases, ranng daa are acqured from mulple sources. Because each source has s own daa dsrbuon ha may be dfferen from ohers, he daa merged from mulple sources are herefore heerogeneous. For example, consder buldng a classfcaon model for oudoor scenes. The ranng mages are colleced from several dfferen ypes of vdeos. Some of he vdeos are news sores and some of hem are of adversemen. Some of hem are of hgh qualy and some of hem are no. Thus, he wdely dsparae characerscs n vdeos cause he merged daa o be heerogeneous. 2) Daa by converng a mulple class problem no a se of bnary class problems. In order o apply he bnary class classfcaon algorhm o mulple class case, we need o Daa Se # Examples #Class # Feaures Ecol Pendg Glass Yeas Vehcle Image/Indoor Image/Oudoor Table 1: Descrpon of daases for he expermen for heerogeneous daa classfcaon. conver he classfcaon problem of mulple classes no a se of bnary class problems. The represenave examples nclude he one-agans-all approach and error correc oupu codng (ECOC) mehod [Deerch, 1995]. Durng hs process, mulple classes are grouped no wo subses of classes. Daa pons from one subse of classes are used as posve examples and he remanng are used as negave examples. Because boh he posve and negave pools can be comprsed of examples from mulple classes, wll creae daa heerogeney for each of he bnary classes. As dscussed, an nuve soluon o classfyng heerogeneous daa s o creae a se of classfcaon models wh each classfer bul on a homogeneous paron (sraum) of he daa, and hen combne classfers for he fnal predcon. The radonal cluserng algorhms are no desgned for hs ask because of he poenal unbalanced cluserszes and he daa fragmenaon problem. Wh he proposed algorhm HISS, we can avod hese wo problems by seng he parameer o be large (0.4 n he expermen). In sum, o classfy heerogeneous daa, we frs apply HISS o oban homogeneous sraa and hen creae a classfcaon model for each sraum o form an ensemble. We wll refer o hs model generaon mehod as HISS-based Model Generaon n our emprcal sudy nex. Fnally, a sackng approach [Wolper, 1992] s used o combne models ha are generaed by he HISS-based model generaon mehod for he fnal predcon of he ensemble. 4. Expermenal Sudy The expermenal sudy s desgned o answer he followng quesons: 1) Is he proposed model generaon mehod effecve for classfyng heerogeneous daa? To hs end, we compare he proposed model generaon mehod o Baggng and AdaBoos n classfyng heerogeneous daases. 2) Is he proposed HISS algorhm effecve for generang relable models? To address hs queson, we wll apply boh he proposed HISS algorhm and he probablsc cluserng algorhm o paron he ranng daa and buld a classfcaon model for each paron. 4.1 Expermenal Desgn Seven dfferen daases are used n he expermens: fve mulple class daases from he UCI Machne Learnng reposory [Blake and Merz, 1998] and wo bnary class daa-
5 Daa Se Baselne AdaBoos Baggng HISS-based Baggng AdaBoos (Sandard) (Sandard) Ensemble (Sackng) (Sackng) Ecol (0.012) (0.006) (0.014) (0.006) (0.006) (0.006) Pendg (0.003) (0.003) (0.002) (0.002) (0.001) (0.003) Glass (0.027) (0.081) (0.046) (0.044) (0.027) (0.027) Yeas (0.012) (0.023) (0.013) (0.013) (0.012) (0.008) Vehcle (0.020) (0.048) (0.024) (0.012) (0.017) (0.033) Image/Indoor (0.008) (0.007) (0.014) (0.013) (0.011) (0.007) Image/Oudoor (0.008) (0.017) (0.011) (0.005) (0.006) (0.007) Table 2: Classfcaon errors for he baselne model (SVM), AdaBoos, Baggng and he propose model generaon mehod ( HISS-based Ensemble ). The column Baggng (Sackng) refers o he case when he ensemble of models s creaed by he Baggng algorhm bu combned hrough he sackng approach usng an SVM. The same s for he column AdaBoos (Sackng). The varance of classfcaon error s lsed n parenhess. ses for mage classfcaon. The characerscs of hese seven daases are lsed n Table 1. For he mulple class daases, we nroduce he heerogeney no he daa by converng he orgnal mulpleclass problem no a bnary one. Smlar o he one-agansall approach, examples from he mos popular class are used as he posve nsances and examples from he remanng classes are assgned o he negave class. Because daa of he negave class are from mulple classes, we would expec some degree of heerogeney nsde he negave class. For he wo daases of mage classfcaon, hey boh are bnary classfcaon problems. The heerogeney of daa s due o he fac ha mages are from seven dfferen vdeo clps and each vdeo clp provdes 500 mages. Snce each vdeo clp s of dfferen ype (e.g., vared qualy n mages), we would expec ceran amoun of heerogeney whn he daa. The baselne algorhm used n hs expermen s suppor vecor machne [Burger, 1998]. In all he expermens, each ensemble mehod generaes 20 dfferen SVMs; a sackng approach [Wolper, 1992] ha also uses a SVM s employed o combne he oupus from all 20 models o form he fnal predcon of he ensemble. For each expermen, we randomly selec 70% of he daa as ranng and he remanng 30% as esng. The expermen s repeaed 10 mes and he average classfcaon error of he en runs s used as he fnal resul wh he varance of classfcaon errors. 4.2 Heerogeneous Daa Classfcaon Table 2 shows classfcaon errors for he baselne suppor Daa Se HISS EM EM (3 Clusers) (10 Clusers) Ecol 0.037(0.006) (0.021) (0.021) Pendg 0.008(0.002) (0.043) (0.023) Glass 0.161(0.044) (0.101) (0.017) Yeas (0.013) (0.013) (0.019) Vehcle (0.012) (0.068) (0.026) Image/Indoor 0.140(0.013) (0.022) (0.014) Image/Oudoor 0.088(0.005) (0.031) (0.036) Table 3: Classfcaon error for usng dfferen cluserng algorhms for model generaon. EM refers o usng Expecaon-Maxmzaon algorhm o cluser daa. vecor machne, he proposed HISS-based ensemble learnng approach, sandard Baggng and sandard AdaBoos. Frs, we can see ha he baselne model performs well comparng wh boh sandard Baggng and AdaBoos. Ths observaon ndcaes ha hese seven heerogeneous daases are raher dffcul for he sandard ensemble approaches o learn. In conras, he proposed HISS-based ensemble mehod performs beer han he baselne model and he wo sandard ensemble mehods. For he daases Glass, Vehcle, and Image/Oudoor, he mprovemen s subsanal, from 38.2% o 16.1% for Galss, 10.3% o 4.8% for Vehcle, and from 11.6% o 8.8% for Image/Oudoor. Snce he HISS-based ensemble mehod uses he sackng approach for combnng dfferen models, s dfferen from he combnaon mehod ha s used by AdaBoos and Baggng. To address hs dfference, we conduc he expermens ha apply a sackng mehod o combne he models generaed by boh Baggng and AdaBoos. The resuls are lsed n Table 2 on he rgh sde of he HISS-based approach, led as Baggng (Sackng) and AdaBoos (Sackng), respecvely. Compared hese resuls o he resuls of Baggng (Sandard) and AdaBoos (Sandard), we see ha here s no subsanal change n classfcaon errors when usng a sackng approach o combne models n ensemble learnng. For all he seven daases, he ensemble of models generaed by HISS performs he bes. The reason why a sackng approach s useful for he HISS-based model generaon mehod bu no o he oher wo s ha models generaed by he HISS-based algorhm are much more dverse han he ones generaed by boh Baggng and AdaBoos. As a resul, applyng anoher layer of classfcaon model o combne he oupus from he dsngushable models (or sackng) wll be able o ake full advanage of all he models and oban he bes performance. Based on he above dscusson, we conclude ha he HISS-based ensemble model s more effecve for classfyng heerogeneous daa han exsng ensemble approaches. 4.3 Comparson wh Oher Cluserng-based Ensemble Mehods The advanage of HISS versus he radonal cluserng algorhms s ha HISS allows each daa pon o be n mulple dfferen sraa. Thus can ensure ha he number of
6 daa pons dsrbued over each sraum s of smlar sze and suffcenly large. In hs expermen, we use boh he radonal cluserng algorhm and he proposed HISS algorhm for model generaon and see how dfferen hey are n classfyng he heerogeneous daases. To observe he effec due o he radeoff beween he number of sraa and he number of daa pons n each sraum, we consder wo dfferen numbers of sraa (or clusers) for he radonal cluserng algorhm: 10 and 3( 1/γ ). We dd no use 20 clusers n he comparson because for some daases he radonal cluserng algorhm s unable o produce he full weny clusers. The radonal cluserng algorhm used n he expermen s he probablsc EM cluserng algorhm. Smlar o he HISSbased ensemble approach, a sackng mehod s used o combne models generaed by he EM cluserng algorhm. The resuls for usng EM cluserng algorhms for model consrucon are lsed n Table 3, led EM (3 clusers) and EM (10 clusers). As suggesed by Table 3, he ncreasng number of clusers can lead o degraded performance. Ths s because a large number of clusers wll form clusers wh a small number of daa pons, whch can be nsuffcen for buldng a relable classfcaon model. On he oher hand, as already ndcaed n he prevous sudy [Derch, 2000], beng able o generae a relavely large number of models s crcal o he success of he ensemble approach. The proposed HISS algorhm can sasfy boh needs by nroducng he subsanal overlappng beween dfferen clusers. As shown n Table 3, he HISS-based mehod ouperforms he EM_cluserng-based ensemble approaches subsanally for almos all daases excep for Yeas (smlar). The mos noceable cases are Ecol and Pendg, for whch he classfcaon errors of EM-based cluserng approaches are one order more han ha of he HISS-based ensemble algorhm. Based on he above expermens and analyss, we conclude ha he HISS-based model generaon s an effecve mehod for model generaon n ensemble learnng for heerogeneous daa classfcaon. 5. Concluson and Fuure Work In hs paper, we propose and examne a new mehod for generang an ensemble of models, whch s o frs paron daa no homogeneous subses and hen creae a model for each subse. A radonal cluserng algorhm lke EM s no suable for he ask of paronng daa due o poenal sze-unbalanced clusers and he daa fragmenaon problem. To address hese wo problems, we propose a novel algorhm HISS, whch allows for daa overlappng beween dfferen clusers (sraa) and promses sze-balanced clusers. Emprcal sudes over seven dfferen heerogeneous daases have shown ha hs new HISS-based model generaon mehod performs very well for heerogeneous daa classfcaon. Currenly, he proposed HISS algorhm assumes equal sze for each sraum (cluser). One possble exenson s o examne alernaves o balance szes of clusers. For example, nsead of enforcng all he clusers o have one sze, we can consran he szes of he clusers no a specfed range o allow some flexbly n mananng hgh homogeney of clusers. References [Hargan and Wong, 1979] Hargan, J.A. and Wong, M.A., A -means Cluserng Algorhm, Appled Sascs 28: [Bremann, 1996] Bremann, L., Baggng Predcaor, Machne Learnng 26, , 1996 [Schapre and Snger, 1999]Schapre, R.E, and Snger, Y., Improved boosng algorhms usng confdence-raed predcons, Machne Learnng 37 (3): , 1999 [Celeux and Govaer, 1992] Celeux, G. and Govaer, G., A Classfcaon EM Algorhm for Cluserng and Two Sochasc Versons, Compuaonal Sascs & Daa Analyss, vol. 14, pp , 1992 [Deerch, 2000] Deerch, T.G., Ensemble Mehods n Machne Learnng. In Mulple Classer Sysems, Caglar, Ialy, 2000 [Dempser e al., 1977] Dempser, A.P., Lard, N.M., and Rubn, D.B., Maxmum Lkelhood from Incomplee Daa va he EM Algorhm, Journal of he Royal sascal Socey, Seres B, 39(1): 1-38, 1977 [Flecher, 1987] Flecher, R., Praccal Mehods of Opmzaon. John Wley and Sons, Inc., 2nd edon, [Deerch and Bakr, 1995] Deerch, T.G. and Bakr, G., Solvng Mulclass Learnng Problems va Error- Correcng Oupu Codes. Journal of Arfcal Inellgence Research, (2): , [Wolper, 1992] Wolper, D.H., Sacked Generalzaon. Neural Neworks, 5: , Pergamon Press, [Burges, 1998] Burges, C.J.C., A Tuoral on Suppor Vecor Machnes for Paern Recognon. Daa Mnng and nowledge Dscovery, 2(2): , [Wen and Frank, 2000] Wen, I.H. and Frank, E., Daa Mnng: Praccal Machne Learnng Tools wh Java Implemenaons. Morgan aufmann, [Blake and Merz, 1998] Blake, C. and Merz, C., UCI reposory of ma-chne learnng daabases. hp:// [Qunlan, 1993] Qunlan, R.J., C4.5: Programs for Machne Learnng, Morgan aufmann, San Maeo, [Qunlan, 1996] Qunlan R.J., Baggng, boosng, and C4.5. In Proceedngs of he Threenh Naonal Conference on Arfcal Inellgence (AAAI 96), [Sh and Malk, 2000] Sh, J., and Malk, J., Normalzed Cu and Image Segmenaon. IEEE Transacons on Paern Analyss and Machne Inellgence, 22(8): , 2000 [Bauer and ohav, 1999] Bauer, E. and ohav, R., An Emprcal Comparson of Vong Classfcaon Algorhms: Baggng, Boosng, and Varans. Machne Learnng, 36(1): , 1999.
Mind the class weight bias: weighted maximum mean discrepancy for unsupervised domain adaptation. Hongliang Yan 2017/06/21
nd he class wegh bas: weghed maxmum mean dscrepancy for unsupervsed doman adapaon Honglang Yan 207/06/2 Doman Adapaon Problem: Tranng and es ses are relaed bu under dfferen dsrbuons. Tranng (Source) DA
More informationUNN: A Neural Network for uncertain data classification
UNN: A Neural Nework for unceran daa classfcaon Jaq Ge, and Yun Xa, Deparmen of Compuer and Informaon Scence, Indana Unversy Purdue Unversy, Indanapols, USA {jaqge, yxa }@cs.upu.edu Absrac. Ths paper proposes
More informationNormal Random Variable and its discriminant functions
Normal Random Varable and s dscrmnan funcons Oulne Normal Random Varable Properes Dscrmnan funcons Why Normal Random Varables? Analycally racable Works well when observaon comes form a corruped sngle prooype
More informationChain-linking and seasonal adjustment of the quarterly national accounts
Sascs Denmark Naonal Accouns 6 July 00 Chan-lnkng and seasonal adjusmen of he uarerly naonal accouns The mehod of chan-lnkng he uarerly naonal accouns was changed wh he revsed complaon of daa hrd uarer
More informationImproving Forecasting Accuracy in the Case of Intermittent Demand Forecasting
(IJACSA) Inernaonal Journal of Advanced Compuer Scence and Applcaons, Vol. 5, No. 5, 04 Improvng Forecasng Accuracy n he Case of Inermen Demand Forecasng Dasuke Takeyasu The Open Unversy of Japan, Chba
More informationLab 10 OLS Regressions II
Lab 10 OLS Regressons II Ths lab wll cover how o perform a smple OLS regresson usng dfferen funconal forms. LAB 10 QUICK VIEW Non-lnear relaonshps beween varables nclude: o Log-Ln: o Ln-Log: o Log-Log:
More informationDeriving Reservoir Operating Rules via Fuzzy Regression and ANFIS
Dervng Reservor Operang Rules va Fuzzy Regresson and ANFIS S. J. Mousav K. Ponnambalam and F. Karray Deparmen of Cvl Engneerng Deparmen of Sysems Desgn Engneerng Unversy of Scence and Technology Unversy
More informationAn Inclusion-Exclusion Algorithm for Network Reliability with Minimal Cutsets
Amercan Journal of ompuaonal Mahemacs, 202, 2, 6-20 hp://dxdoorg/0426/acm2022404 Publshed Onlne December 202 (hp://wwwscrporg/ournal/acm) An Incluson-Excluson Algorhm for ework Relably wh Mnmal uses Yan-Ru
More informationSection 6 Short Sales, Yield Curves, Duration, Immunization, Etc.
More Tuoral a www.lledumbdocor.com age 1 of 9 Secon 6 Shor Sales, Yeld Curves, Duraon, Immunzaon, Ec. Shor Sales: Suppose you beleve ha Company X s sock s overprced. You would ceranly no buy any of Company
More informationA valuation model of credit-rating linked coupon bond based on a structural model
Compuaonal Fnance and s Applcaons II 247 A valuaon model of cred-rang lnked coupon bond based on a srucural model K. Yahag & K. Myazak The Unversy of Elecro-Communcaons, Japan Absrac A cred-lnked coupon
More informationSOCIETY OF ACTUARIES FINANCIAL MATHEMATICS. EXAM FM SAMPLE SOLUTIONS Interest Theory
SOCIETY OF ACTUARIES EXAM FM FINANCIAL MATHEMATICS EXAM FM SAMPLE SOLUTIONS Ineres Theory Ths page ndcaes changes made o Sudy Noe FM-09-05. January 4, 04: Quesons and soluons 58 60 were added. June, 04
More informationEstimation of Optimal Tax Level on Pesticides Use and its
64 Bulgaran Journal of Agrculural Scence, 8 (No 5 0, 64-650 Agrculural Academy Esmaon of Opmal Ta Level on Pescdes Use and s Impac on Agrculure N. Ivanova,. Soyanova and P. Mshev Unversy of Naonal and
More informationOnline Technical Appendix: Estimation Details. Following Netzer, Lattin and Srinivasan (2005), the model parameters to be estimated
Onlne Techncal Appendx: Esmaon Deals Followng Nezer, an and Srnvasan 005, he model parameers o be esmaed can be dvded no hree pars: he fxed effecs governng he evaluaon, ncdence, and laen erence componens
More informationAccuracy of the intelligent dynamic models of relational fuzzy cognitive maps
Compuer Applcaons n Elecrcal Engneerng Accuracy of he nellgen dynamc models of relaonal fuzzy cognve maps Aleksander Jasrebow, Grzegorz Słoń Kelce Unversy of Technology 25-314 Kelce, Al. Tysącleca P. P.
More informationFITTING EXPONENTIAL MODELS TO DATA Supplement to Unit 9C MATH Q(t) = Q 0 (1 + r) t. Q(t) = Q 0 a t,
FITTING EXPONENTIAL MODELS TO DATA Supplemen o Un 9C MATH 01 In he handou we wll learn how o fnd an exponenal model for daa ha s gven and use o make predcons. We wll also revew how o calculae he SSE and
More informationQuarterly Accounting Earnings Forecasting: A Grey Group Model Approach
Quarerly Accounng Earnngs Forecasng: A Grey Group Model Approach Zheng-Ln Chen Deparmen of Accounng Zhongnan Unversy of Economcs and Law # Souh Nanhu Road, Wuhan Cy, 430073 Hube People's Republc of Chna
More informationAnalysing Big Data to Build Knowledge Based System for Early Detection of Ovarian Cancer
Indan Journal of Scence and Technology, Vol 8(4), DOI: 0.7485/js/205/v84/65745, July 205 ISSN (Prn) : 0974-6846 ISSN (Onlne) : 0974-5645 Analysng Bg Daa o Buld Knowledge Based Sysem for Early Deecon of
More informationDifferences in the Price-Earning-Return Relationship between Internet and Traditional Firms
Dfferences n he Prce-Earnng-Reurn Relaonshp beween Inerne and Tradonal Frms Jaehan Koh Ph.D. Program College of Busness Admnsraon Unversy of Texas-Pan Amercan jhkoh@upa.edu Bn Wang Asssan Professor Compuer
More informationExplaining Product Release Planning Results Using Concept Analysis
Explanng Produc Release Plannng Resuls Usng Concep Analyss Gengshen Du, Thomas Zmmermann, Guenher Ruhe Deparmen of Compuer Scence, Unversy of Calgary 2500 Unversy Drve NW, Calgary, Albera T2N 1N4, Canada
More informationEconomics of taxation
Economcs of axaon Lecure 3: Opmal axaon heores Salane (2003) Opmal axes The opmal ax sysem mnmzes he excess burden wh a gven amoun whch he governmen wans o rase hrough axaon. Opmal axes maxmze socal welfare,
More informationAmerican basket and spread options. with a simple binomial tree
Amercan baske and spread opons wh a smple bnomal ree Svelana orovkova Vre Unverse Amserdam Jon work wh Ferry Permana acheler congress, Torono, June 22-26, 2010 1 Movaon Commody, currency baskes conss of
More informationTax Dispute Resolution and Taxpayer Screening
DISCUSSION PAPER March 2016 No. 73 Tax Dspue Resoluon and Taxpayer Screenng Hdek SATO* Faculy of Economcs, Kyushu Sangyo Unversy ----- *E-Mal: hsao@p.kyusan-u.ac.jp Tax Dspue Resoluon and Taxpayer Screenng
More informationNetwork Security Risk Assessment Based on Node Correlation
Journal of Physcs: Conference Seres PAPER OPE ACCESS ewor Secury Rs Assessmen Based on ode Correlaon To ce hs arcle: Zengguang Wang e al 2018 J. Phys.: Conf. Ser. 1069 012073 Vew he arcle onlne for updaes
More informationOptimal Combination of Trading Rules Using Neural Networks
Vol. 2, No. Inernaonal Busness Research Opmal Combnaon of Tradng Rules Usng Neural Neworks Subraa Kumar Mra Professor, Insue of Managemen Technology 35 Km Mlesone, Kaol Road Nagpur 44 502, Inda Tel: 9-72-280-5000
More informationAn improved segmentation-based HMM learning method for Condition-based Maintenance
An mproved segmenaon-based HMM learnng mehod for Condon-based Manenance T Lu 1,2, J Lemere 1,2, F Carella 1,2 and S Meganck 1,3 1 ETRO Dep., Vre Unverse Brussel, Plenlaan 2, 1050 Brussels, Belgum 2 FMI
More informationA Neural Network Approach to Time Series Forecasting
A Neural Nework Approach o Tme Seres Forecasng Iffa A. Gheyas, Lesle S. Smh Absrac We propose a smple approach for forecasng unvarae me seres. The proposed algorhm s an ensemble learnng echnque ha combnes
More informationDynamic Relationship and Volatility Spillover Between the Stock Market and the Foreign Exchange market in Pakistan: Evidence from VAR-EGARCH Modelling
Dynamc Relaonshp and Volaly pllover Beween he ock Marke and he Foregn xchange marke n Paksan: vdence from VAR-GARCH Modellng Dr. Abdul Qayyum Dr. Muhammad Arshad Khan Inroducon A volale sock and exchange
More informationNoise and Expected Return in Chinese A-share Stock Market. By Chong QIAN Chien-Ting LIN
Nose and Expeced Reurn n Chnese A-share Sock Marke By Chong QIAN Chen-Tng LIN 1 } Capal Asse Prcng Model (CAPM) by Sharpe (1964), Lnner (1965) and Mossn (1966) E ( R, ) R f, + [ E( Rm, ) R f, = β ] + ε
More informationOnline Adaboost-Based Parameterized Methods for Dynamic Distributed Network Intrusion Detection
Onlne Adaboos-Based Parameerzed Mehods or Dnamc Dsrbued Nework Inruson Deecon Wemng Hu, Jun Gao, Yanguo Wang, and Ou Wu (Naonal Laboraor o Paern Recognon, Insue o Auomaon, Chnese Academ o Scences, Beng
More informationBaoding, Hebei, China. *Corresponding author
2016 3 rd Inernaonal Conference on Economcs and Managemen (ICEM 2016) ISBN: 978-1-60595-368-7 Research on he Applcably of Fama-French Three-Facor Model of Elecrc Power Indusry n Chnese Sock Marke Yeld
More informationA Hybrid Method to Improve Forecasting Accuracy Utilizing Genetic Algorithm An Application to the Data of Operating equipment and supplies
A Hyrd Mehod o Improve Forecasng Accuracy Ulzng Genec Algorhm An Applcaon o he Daa of Operang equpmen and supples Asam Shara Tax Corporaon Arkne, Shzuoka Cy, Japan, e-mal: a-shara@arkne.nfo Dasuke Takeyasu
More informationComparing Sharpe and Tint Surplus Optimization to the Capital Budgeting Approach with Multiple Investments in the Froot and Stein Framework.
Comparng Sharpe and Tn Surplus Opmzaon o he Capal Budgeng pproach wh Mulple Invesmens n he Froo and Sen Framework Harald Bogner Frs Draf: Sepember 9 h 015 Ths Draf: Ocober 1 h 015 bsrac Below s shown ha
More informationOnline appendices from Counterparty Risk and Credit Value Adjustment a continuing challenge for global financial markets by Jon Gregory
Onlne appendces fro Counerpary sk and Cred alue Adusen a connung challenge for global fnancal arkes by Jon Gregory APPNDX A: Dervng he sandard CA forula We wsh o fnd an expresson for he rsky value of a
More informationDEA-Risk Efficiency and Stochastic Dominance Efficiency of Stock Indices *
JEL Classfcaon: C61, D81, G11 Keywords: Daa Envelopmen Analyss, rsk measures, ndex effcency, sochasc domnance DEA-Rsk Effcency and Sochasc Domnance Effcency of Sock Indces * Marn BRANDA Charles Unversy
More informationFugit (options) The terminology of fugit refers to the risk neutral expected time to exercise an
Fug (opons) INTRODUCTION The ermnology of fug refers o he rsk neural expeced me o exercse an Amercan opon. Invened by Mark Garman whle professor a Berkeley n he conex of a bnomal ree for Amercan opon hs
More informationCorrelation of default
efaul Correlaon Correlaon of defaul If Oblgor A s cred qualy deeroraes, how well does he cred qualy of Oblgor B correlae o Oblgor A? Some emprcal observaons are efaul correlaons are general low hough hey
More informationANFIS Based Time Series Prediction Method of Bank Cash Flow Optimized by Adaptive Population Activity PSO Algorithm
Informaon 25, 6, 3-33; do:.339/nfo633 Arcle OPEN ACCESS nformaon ISSN 278-2489 www.mdp.com/journal/nformaon ANFIS Based Tme Seres Predcon Mehod of Bank Cash Flow Opmzed by Adapve Populaon Acvy PSO Algorhm
More informationMixtures of Normal Distributions: Application to Bursa Malaysia Stock Market Indices
World Appled Scences Journal 6 (6): 78-790, 0 ISSN 88-495 IDOSI Publcaons, 0 Mxures of Normal Dsrbuons: Applcaon o Bursa Malaysa Sock Marke Indces Zey An Kamaruzzaman, Zad Isa and Mohd ahr Ismal School
More informationA Novel Particle Swarm Optimization Approach for Grid Job Scheduling
A Novel Parcle warm Opmzaon Approach for Grd ob chedulng Hesam Izaan, Behrouz Tor Ladan, Kamran Zamanfar, Ajh Abraham³ Islamc Azad Unversy, Ramsar branch, Ramsar, Iran zaan@gmal.com Deparmen of Compuer
More informationCryptographic techniques used to provide integrity of digital content in long-term storage
RB/3/2011 Crypographc echnques used o provde negry of dgal conen n long-erm sorage REPORT ON THE PROBLEM Problem presened by Marn Šmka Paweł Wojcechowsk Polsh Secury Prnng Works (PWPW) 1 Repor auhors Małgorzaa
More informationThe UAE UNiversity, The American University of Kurdistan
MPRA Munch Personal RePEc Archve A MS-Excel Module o Transform an Inegraed Varable no Cumulave Paral Sums for Negave and Posve Componens wh and whou Deermnsc Trend Pars. Abdulnasser Haem-J and Alan Musafa
More informationMichał Kolupa, Zbigniew Śleszyński SOME REMARKS ON COINCIDENCE OF AN ECONOMETRIC MODEL
M I S C E L L A N E A Mchał Kolupa, bgnew Śleszyńsk SOME EMAKS ON COINCIDENCE OF AN ECONOMETIC MODEL Absrac In hs paper concep of concdence of varable and mehods for checkng concdence of model and varables
More informationCointegration between Fama-French Factors
1 Conegraon beween Fama-French Facors Absrac Conegraon has many applcaons n fnance and oher felds of scence researchng me seres and her nerdependences. The analyss s a useful mehod o analyse non-conegraon
More informationFairing of Polygon Meshes Via Bayesian Discriminant Analysis
Farng of Polygon Meshes Va Bayesan Dscrmnan Analyss Chun-Yen Chen Insue of Informaon Scence, Academa Snca. Deparmen of Compuer Scence and Informaon Engneerng, Naonal Tawan Unversy. 5, Tawan, Tape, Nankang
More informationLearning From Labeled And Unlabeled Data: An Empirical Study Across Techniques And Domains
Journal of Arfcal Inellgence Research 3 (005 33 -- 366 Submed 06/04; publshed 03/05 Learnng From Labeled And Unlabeled Daa: An Emprcal Sud Across Technques And Domans Nesh V. Chawla Deparmen of Compuer
More informationFloating rate securities
Caps and Swaps Floang rae secures Coupon paymens are rese perodcally accordng o some reference rae. reference rae + ndex spread e.g. -monh LIBOR + 00 bass pons (posve ndex spread 5-year Treasury yeld 90
More informationThe Empirical Research of Price Fluctuation Rules and Influence Factors with Fresh Produce Sequential Auction Limei Cui
6h Inernaonal Conference on Sensor Nework and Compuer Engneerng (ICSNCE 016) The Emprcal Research of Prce Flucuaon Rules and Influence Facors wh Fresh Produce Sequenal Aucon Lme Cu Qujng Normal Unversy,
More informationPrediction of Oil Demand Based on Time Series Decomposition Method Nan MA * and Yong LIU
2017 2nd Inernaonal Conference on Sofware, Mulmeda and Communcaon Engneerng (SMCE 2017) ISBN: 978-1-60595-458-5 Predcon of Ol Demand Based on Tme Seres Decomposon Mehod Nan MA * and Yong LIU College of
More informationMULTI-SPECTRAL IMAGE ANALYSIS BASED ON DYNAMICAL EVOLUTIONARY PROJECTION PURSUIT
MULTI-SPECTRAL IMAGE AALYSIS BASED O DYAMICAL EVOLUTIOARY PROJECTIO PURSUIT YU Changhu a, MEG Lngku a, YI Yaohua b, a School of Remoe Sensng Informaon Engneerng, Wuhan Unversy, 39#,Luoyu Road, Wuhan,Chna,430079,
More informationAgricultural and Rural Finance Markets in Transition
Agrculural and Rural Fnance Markes n Transon Proceedngs of Regonal Research Commee NC-04 S. Lous, Mssour Ocober 4-5, 007 Dr. Mchael A. Gunderson, Edor January 008 Food and Resource Economcs Unversy of
More informationBatch Processing for Incremental FP-tree Construction
Inernaonal Journal of Compuer Applons (975 8887) Volume 5 No.5, Augus 21 Bach Processng for Incremenal FP-ree Consrucon Shashkumar G. Toad Deparmen of CSE, GMRIT, Rajam, Srkakulam Dsrc AndraPradesh, Inda.
More informationVI. Clickstream Big Data and Delivery before Order Making Mode for Online Retailers
VI. Clcksream Bg Daa and Delvery before Order Makng Mode for Onlne Realers Yemng (Yale) Gong EMLYON Busness School Haoxuan Xu *, Jnlong Zhang School of Managemen, Huazhong Unversy of Scence &Technology
More informationOptimal Fuzzy Min-Max Neural Network (FMMNN) for Medical Data Classification Using Modified Group Search Optimizer Algorithm
1 Opmal Fuzzy Mn-Max Neural Nework (FMMNN) for Medcal Daa Classfcaon Usng Modfed Group Search Opmzer Algorhm D. Mahammad Raf 1 * Chear Ramachandra Bharah 2 1 Vvekananda Insue of Engneerng & Technology,
More informationIFX-Cbonds Russian Corporate Bond Index Methodology
Approved a he meeng of he Commee represenng ZAO Inerfax and OOO Cbonds.ru on ovember 1 2005 wh amendmens complan wh Agreemen # 545 as of ecember 17 2008. IFX-Cbonds Russan Corporae Bond Index Mehodology
More informationImproving Earnings per Share: An Illusory Motive in Stock Repurchases
Inernaonal Journal of Busness and Economcs, 2009, Vol. 8, No. 3, 243-247 Improvng Earnngs per Share: An Illusory Move n Sock Repurchases Jong-Shn We Deparmen of Inernaonal Busness Admnsraon, Wenzao Ursulne
More informationInterest Rate Derivatives: More Advanced Models. Chapter 24. The Two-Factor Hull-White Model (Equation 24.1, page 571) Analytic Results
Ineres Rae Dervaves: More Advanced s Chaper 4 4. The Two-Facor Hull-Whe (Equaon 4., page 57) [ θ() ] σ 4. dx = u ax d dz du = bud σdz where x = f () r and he correlaon beween dz and dz s ρ The shor rae
More informationEstimating intrinsic currency values
Esmang nrnsc currency values Forex marke praconers consanly alk abou he srenghenng or weakenng of ndvdual currences. In hs arcle, Jan Chen and Paul Dous presen a new mehodology o quanfy hese saemens n
More informationA Hybrid Method for Forecasting with an Introduction of a Day of the Week Index to the Daily Shipping Data of Sanitary Materials
Journal of Communcaon and Compuer (05) 0-07 do: 0.765/548-7709/05.0.00 D DAVID PUBLISHING A Hyrd Mehod for Forecasng wh an Inroducon of a Day of he Week Inde o he Daly Shppng Daa of Sanary Maerals Dasuke
More informationRecursive Data Mining for Masquerade Detection and Author Identification
Recursve Daa Mnng for Masquerade Deecon and Auhor Idenfcaon Boleslaw K. Szymansk, IEEE Fellow, and Yongqang Zhang Deparmen of Compuer Scence, RPI, Troy, NY 280, USA Absrac- In hs paper, a novel recursve
More informationMethodology of the CBOE S&P 500 PutWrite Index (PUT SM ) (with supplemental information regarding the CBOE S&P 500 PutWrite T-W Index (PWT SM ))
ehodology of he CBOE S&P 500 PuWre Index (PUT S ) (wh supplemenal nformaon regardng he CBOE S&P 500 PuWre T-W Index (PWT S )) The CBOE S&P 500 PuWre Index (cker symbol PUT ) racks he value of a passve
More informationThe Proposed Mathematical Models for Decision- Making and Forecasting on Euro-Yen in Foreign Exchange Market
Iranan Economc Revew, Vol.6, No.30, Fall 20 The Proposed Mahemacal Models for Decson- Makng and Forecasng on Euro-Yen n Foregn Exchange Marke Abdorrahman Haer Masoud Rabban Al Habbna Receved: 20/07/24
More informationAssociating Absent Frequent Itemsets with Infrequent Items to Identify Abnormal Transactions
Assocang Absen Frequen Iemses wh Infrequen Iems o Idenfy Abnormal Transacons L-Jen Kao Deparmen of Compuer Scence and Informaon Engneerng Hwa Hsa Insue of Technology New Tape Cy, Tawan 23568 ljenkao@cc.hwh.edu.w
More informationBank of Japan. Research and Statistics Department. March, Outline of the Corporate Goods Price Index (CGPI, 2010 base)
Bank of Japan Research and Sascs Deparmen Oulne of he Corporae Goods Prce Index (CGPI, 2010 base) March, 2015 1. Purpose and Applcaon The Corporae Goods Prce Index (CGPI) measures he prce developmens of
More informationSome Insights of Value-Added Tax Gap
Ovdus Unversy Annals, Economc Scences Seres Some Insghs of Value-Added Tax Ga Cuceu Ionuţ-Consann Vădean Vorela-Lga Maşca Smona-Gabrela "Babeş-Bolya" Unversy Cluj-Naoca, Faculy of Economcs and Busness
More informationCo-Integration Study of Relationship between Foreign Direct Investment and Economic Growth
www.ccsene.org/br Inernaonal Busness Research Vol. 4, No. 4; Ocober 2011 Co-Inegraon Sudy of Relaonshp beween Foregn Drec Invesen and Econoc Growh Haao Sun Qngdao Technologcal Unversy, Qngdao 266520, Chna
More informationTruth Discovery in Data Streams: A Single-Pass Probabilistic Approach
Truh Dscovery n Daa Sreams: A Sngle-Pass Probablsc Approach Zhou Zhao, James Cheng and Wlfred Ng Deparmen of Compuer Scence and Engneerng, Hong Kong Unversy of Scence and Technology Deparmen of Compuer
More informationPattern Classification (V) 杜俊
Paern Cassfcaon V 杜俊 jundu@usc.edu.cn Oune Bayesan Decson heory How o mae he oma decson? Maxmum a oseror MAP decson rue Generave Modes Jon dsrbuon of observaon and abe sequences Mode esmaon: MLE Bayesan
More informationThe Financial System. Instructor: Prof. Menzie Chinn UW Madison
Economcs 435 The Fnancal Sysem (2/13/13) Insrucor: Prof. Menze Chnn UW Madson Sprng 2013 Fuure Value and Presen Value If he presen value s $100 and he neres rae s 5%, hen he fuure value one year from now
More informationA Multi-Periodic Optimization Modeling Approach for the Establishment of a Bike Sharing Network: a Case Study of the City of Athens
A Mul-Perodc Opmzaon Modelng Approach for he Esablshmen of a Be Sharng Newor: a Case Sudy of he Cy of Ahens G.K.D Sahards, A. Fragogos and E. Zygour Absrac Ths sudy nroduces a novel mahemacal formulaon
More informationHas the Euro Era Facilitated Inflation Convergence?
Holmes, Journal of Inernaonal and Global Economc Sudes, 1(1), June 2008, 27-41 27 Has he Euro Era Faclaed Inflaon Convergence? Mark J. Holmes Unversy of Wakao, New Zealand Absrac Ths paper nvesgaes he
More informationSUMMARY INTRODUCTION. Figure 1: An illustration of the integration of well log data and seismic data in a survey area. Seismic cube. Well-log.
Perophyscal propery esmaon from sesmc daa usng recurren neural neworks Moaz Alfarraj, and Ghassan AlRegb Cener for Energy and Geo Processng CeGP, Georga Insue of Technology SUMMARY Reservor characerzaon
More informationA Change Detection Model for Credit Card Usage Behavior
Proceedngs of he 5h WSEAS In. Conf. on COMPUTATIONAL INTELLIGENCE, MAN-MACHINE SYSTEMS AND CYBERNETICS, Vence, Ialy, November 20-22, 2006 276 A Change Deecon Model for Cred Card Usage Behavor CHIEH-YUAN
More informationA Framework for Large Scale Use of Scanner Data in the Dutch CPI
A Framework for Large Scale Use of Scanner Daa n he Duch CPI Jan de Haan Sascs Neherlands and Delf Unversy of Technology Oawa Group, 2-22 May 215 The basc dea Ideally, o make he producon process as effcen
More informationThe Effects of Nature on Learning in Games
The Effecs of Naure on Learnng n Games C.-Y. Cynha Ln Lawell 1 Absrac Ths paper develops an agen-based model o nvesgae he effecs of Naure on learnng n games. In parcular, I exend one commonly used learnng
More informationOnline Data, Fixed Effects and the Construction of High-Frequency Price Indexes
Onlne Daa, Fxed Effecs and he Consrucon of Hgh-Frequency Prce Indexes Jan de Haan* and Rens Hendrks** * ascs eherlands / Delf Unversy of Technology ** ascs eherlands EMG Worksho 23 Ams of he aer Exlan
More informationThe Keynesian micro-foundations of the business cycle: some implications of globalisation
The Keynesan mcro-foundaons of he busness cycle: some mplcaons of globalsaon Paul Ormerod, Volerra Consulng Ld., London e-mal pormerod@volerra.co.uk el: 44 0208 878 6333 I am graeful o Rod Gbson n parcular
More informationHardware-Assisted High-Efficiency Ray Casting of Unstructured Time-Varying Flows Using Temporal Coherence
Hardware-Asssed Hgh-Effcency Ray Casng of Unsrucured Tme-Varyng Flows Usng Temporal Coherence Qanl Ma, Lang Zeng, Huaxun Xu, Wenke Wang, Skun L Absrac Advances n compuaonal power are enablng hgh-precson
More informationA Novel Application of the Copula Function to Correlation Analysis of Hushen300 Stock Index Futures and HS300 Stock Index
A Novel Applcaon of he Copula Funcon o Correlaon Analyss of Hushen3 Sock Index Fuures and HS3 Sock Index Fang WU *, 2, Yu WEI. School of Economcs and Managemen, Souhwes Jaoong Unversy, Chengdu 63, Chna
More informationGaining From Your Own Default
Ganng From Your Own Defaul Jon Gregory jon@ofranng.com Jon Gregory (jon@ofranng.com), Quan ongress US, 14 h July 2010 page 1 Regulaon s Easy () Wha don lke as a regulaor? Dfferen nsuons valung asses dfferenly
More informationConditional Skewness of Aggregate Market Returns: Evidence from Developed and Emerging Markets
Global Economy and Fnance Journal Vol. 7. No.. March 04. Pp. 96 Condonal Skewness of Aggregae Marke Reurns: Evdence from Developed and Emergng Markes Anchada Charoenrook and Hazem Daouk Ths paper examnes
More informationAdjusted-Productivity Growth for Resource Rents: Kuwait Oil Industry
Appled Economcs and Fnance Vol. 3, No. 2; May 2016 ISSN 2332-7294 E-ISSN 2332-7308 Publshed by Redfame Publshng URL: hp://aef.redfame.com Adjused-Producvy Growh for Resource Rens: Kuwa Ol Indusry 1 Acng
More informationImpact of Stock Markets on Economic Growth: A Cross Country Analysis
Impac of Sock Markes on Economc Growh: A Cross Counry Analyss By Muhammad Jaml Imporance of sock markes for poolng fnancal resources ncreased snce he las wo decades. Presen sudy analyzed mpac of sock markes
More informationUC San Diego Recent Work
UC San Dego Recen Work Tle On More Robus Esmaon of Skewness and Kuross: Smulaon and Applcaon o he S&P500 Index Permalnk hps://escholarshp.org/uc/em/7b5v07p Auhors Km, Tae-Hwan Whe, Halber Publcaon Dae
More informationNumerical Evaluation of European Option on a Non Dividend Paying Stock
Inernaonal Journal of Compuaonal cence and Mahemacs. IN 0974-389 olume Number 3 (00) pp. 6--66 Inernaonal Research Publcaon House hp://www.rphouse.com Numercal Evaluaon of European Opon on a Non Dvdend
More informationCareer wage profiles and the minimum wage
Career wage profles and he mnmum wage Kerry L. Papps A model of on-he-job ranng n he presence of a mnmum wage s presened. Ths predcs ha, n mos cases, he mnmum wage wll have a negave effec on a worker s
More informationThe Selection Ability of Italian Mutual Fund. By Valter Lazzari and Marco Navone
The Selecon Ably of Ialan Muual Fund By Valer Lazzar and Marco Navone Workng Paper N. 1/3 Ocober 23 THE SELECTION ABILITY OF ITALIAN MUTUAL FUND MANAGERS By Valer Lazzar Professor of Bankng and Fnance
More informationUnified Unit Commitment Formulation and Fast Multi-Service LP Model for Flexibility Evaluation in Sustainable Power Systems
IEEE Transacons on Susanable Energy Acceped for publcaon, November 2015 1 Unfed Un Commmen Formulaon and Fas Mul-Servce LP Model for Flexbly Evaluaon n Susanable Power Sysems Lngx Zhang, Suden Member,
More informationTrade Between Euro Zone and Arab Countries: a Panel Study. By Nasri HARB* United Arab Emirates University Department of Economics P.O.
Trade Beween Euro Zone and Arab Counres: a Panel Sudy By Nasr HARB* Uned Arab Emraes Unversy Deparmen of Economcs P.O. Box 17555, Al-An, Uned Arab Emraes nasr.harb@uaeu.ac.ae Ocober 2005 Absrac We consruc
More informationInteractive Dynamic Influence Diagrams
Ineracve Dynamc Influence Dagrams Kyle Polch and Por Gmyrasewcz Deparmen of Compuer Scence, Unversy of Illnos a Chcago Chcago, IL, 60607-7053, USA E-mal: kpolch@cs.uc.edu, por@cs.uc.edu Absrac Ths paper
More informationKeywords: School bus problem, heuristic, harmony search
Journal of Emergng Trends n Compung and Informaon Scences 2009-2013 CIS Journal. All rghs reserved. hp://www.csjournal.org Model and Algorhm for Solvng School Bus Problem 1 Taehyeong Km, 2 Bum-Jn Par 1
More informationPermanent Income and Consumption
roceedngs of 30h Inernaonal onference Mahemacal Mehods n Economcs ermanen Income and onsumpon Václava ánková 1 Absrac. A heory of consumer spendng whch saes ha people wll spend money a a level conssen
More informationDYNAMIC ECONOMETRIC MODELS Vol. 8 Nicolaus Copernicus University Toruń 2008
DYNAMIC ECONOMETRIC MODELS Vol. 8 Ncolaus Coperncus Unversy Toruń 2008 Por Fszeder Ncolaus Coperncus Unversy n Toruń Julusz Preś Szczecn Unversy of Technology Prcng of Weaher Opons for Berln Quoed on he
More informationVolatility Modeling for Forecasting Stock Index with Fixed Parameter Distributional Assumption
Journal of Appled Fnance & Banng, vol. 3, no. 1, 13, 19-1 ISSN: 179-5 (prn verson), 179-599 (onlne) Scenpress Ld, 13 Volaly Modelng for Forecasng Soc Index wh Fxed Parameer Dsrbuonal Assumpon Md. Mosafzur
More informationForecasting Value-at-Risk under Different Distributional Assumptions
Arcle Forecasng Value-a-Rsk under Dfferen Dsrbuonal Assumpons Manuela Braone 1, and Ncolas K. Scholes 1,,, Receved: 1 July 1; Acceped: 1 December 1; Publshed: 11 January 16 Academc Edors: Fredj Jawad,
More informationDecision Support for Service Transition Management
Decson Suppor for Servce Transon Managemen Enforce Change Schedulng by Performng Change Rsk and Busness Impac Analyss Thomas Sezer Technsche Unversä München Char of Inerne-based Informaon Sysems 85748
More informationConditional Skewness of Aggregate Market Returns
Condonal Skewness of Aggregae Marke Reurns Anchada Charoenrook and Hazem Daouk + March 004 Ths verson: May 008 Absrac: The skewness of he condonal reurn dsrbuon plays a sgnfcan role n fnancal heory and
More informationESSAYS ON MONETARY POLICY AND INTERNATIONAL TRADE. A Dissertation HUI-CHU CHIANG
ESSAYS ON MONETARY POLICY AND INTERNATIONAL TRADE A Dsseraon by HUI-CHU CHIANG Submed o he Offce of Graduae Sudes of Texas A&M Unversy n paral fulfllmen of he requremens for he degree of DOCTOR OF PHILOSOPHY
More informationSTOCK PRICES TEHNICAL ANALYSIS
STOCK PRICES TEHNICAL ANALYSIS Josp Arnerć, Elza Jurun, Snježana Pvac Unversy of Spl, Faculy of Economcs Mace hrvaske 3 2 Spl, Croaa jarnerc@efs.hr, elza@efs.hr, spvac@efs.hr Absrac Ths paper esablshes
More informationUsing Fuzzy-Delphi Technique to Determine the Concession Period in BOT Projects
Usng Fuzzy-Delph Technque o Deermne he Concesson Perod n BOT Projecs Khanzad Mosafa Iran Unversy of Scence and Technology School of cvl engneerng Tehran, Iran. P.O. Box: 6765-63 khanzad@us.ac.r Nasrzadeh
More informationA New Method to Measure the Performance of Leveraged Exchange-Traded Funds
A ew Mehod o Measure he Performance of Leveraged Exchange-Traded Funds Ths verson: Sepember 03 ara Charupa DeGrooe School of Busness McMaser Unversy 80 Man Sree Wes Hamlon, Onaro L8S 4M4 Canada Tel: (905)
More information