Ho June 11 2003 Abstract This article discusses the use of symmetric ultiplicativ in teraction e57355ect to capture cer tain yp es of thirdorder dep endence patterns often presen in so cial net orks and other dy adic datasets Suc an e57355ect along ID: 80878
Download Pdf The PPT/PDF document "Bilinear Mixed Eects Mo dels for Dy adic..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
BilinearMixedEectsModelsforDyadicDataPeterD.HoJune11,2003AbstractThisarticlediscussestheuseofasymmetricmultiplicativeinteractioneecttocapturecer-taintypesofthird-orderdependencepatternsoftenpresentinsocialnetworksandotherdyadicdatasets.Suchaneect,alongwithstandardlinearxedandrandomeects,isincorporatedintoageneralizedlinearmodel,andaMarkovchainMonteCarloalgorithmisprovidedforBayesianestimationandinference.Inanexampleanalysisofinternationalrelationsdata,accountingforsuchpatternsimprovesmodeltandpredictiveperformance.KEYWORDS:socialnetwork,balance,innerproductscaling,generalizedlinearmodel.1IntroductionDyadicdataconsistofmeasurementsthataremadeonpairsofobjectsorunderapairofconditions,sothatyi;jdenotesthevalueofthe(possiblydirected)measurementfromitoj.Examplesincludesocialnetworkanalysis,\roundrobin"experimentsinpsychology,andcomparativedatainwhichyi;jmightbeameasureofsimilaritybetweenunitsiandj.Inthesocialnetworksliterature,modelinghasfocusedonthebinarycasewhereyi;jiseitherzeroorone,indicatingthepresenceorabsenceofa\link"fromitoj.Thishasledtothedevelopmentofdataanalysistoolsbasedondirectedgraphsandthestudyofexponentiallyparameterizedrandomgraphmodels(WassermanandPattison1996).Forvalued(non-binary)dyadicdatasets,aperceivedlackofstatisticaltoolshassometimesledtoad-hocreductionsofvaluedresponsestobinarydata.However,ANOVAmethodsareavailableforvalueddyadicdata:theso-calledsocialrelationsmodel(Warner,Kenny,andStoto1979;Wong1982)allowsforthedecompositionofthevarianceintosenderandreceiverspeciceects,aswellasallowsforcorrelationbetweenresponseswithinadyad.SuchamodelhasbeenstudiedinthecontextofalineargroupsymmetrymodelbyLi(2002),andadvancesinvariancecomponentanalysishavebeenmadebyandGillandSwartz(2001)andLiandLokenPeterD.HoisAssistantProfessorofStatistics,Box354322,UniversityofWashington,SeattleWA98195-4322,Email:ho@stat.washington.edu,Web:www.stat.washington.edu/ho.ThisresearchwassupportedbyOceofNavalResearchgrantN00014-02-1-1011.TheauthorthanksMarkHandcockandMichaelWardforhelpfuldiscussions.1 (2002).Thesemodelsgenerallypresumenormallydistributeddataandadditiveeects,andthusthelackofanysortofdependencebeyondthosespeciedbysecond-ordermoments.Incontrast,manyobserveddyadicdatasetsexhibitcertainformsofthird-orderdependence,andoftenitisofscienticinteresttoquantifythesehigherorderpatterns.Inthisarticleweproposeaclassofgeneralizedadditivemodelsbasedonthesocialrelationsmodel,butincorporatethirdorderdependenceviaabilineareect.Thebilineareectforapair(i;j)issimplytheinnerproductofunobservedcharacteristicvectorsziandzj,specictounitsiandjrespectively.ThisapproachissimilarinspirittothelatentvariablemethodsproposedbyHo,Raftery,andHandcock(2002)tocapturetransitivityinasocialnetworkdataset,buthassomecomputationalandconceptualadvantages.Thebilineareectisalsoatypeofmultiplicativeinteraction(Gabriel1978;MarasingheandJohnson1982;Oman1991).ThemodelspresentedinthisarticlearesimilartothegeneralizedbilinearregressionmodelsstudiedbyGabriel(1998),whoconsideredapproximatemaximumlikelihoodestimationinthecontextoffactorialdesigns.Inthisarticle,weshowhowabilineareectcanbeusedtorepresentcertainformsofdependenceoftenseenindyadicdata,anddevelopaMarkovchainMonteCarloalgorithmbasedonGibbssampling,providingarbitrarilyexactBayesianinference.Withsomemodications,thealgorithmcanbeusedasameansofmakingBayesianinferenceforabroadclassofgeneralizedbilinearregressionmodelswithmixedeects.Inthenextsection,wediscussthebasiclinearmixedeectsmodelfordyadicdataandtheresultingdependencestructure.InSection3,wediscusstypesofthird-orderdependenceoftenseeninnetworkdatasetsandtheuseofabilineareecttocapturesuchdependence.Section4givesaMarkovchainMonteCarlo(MCMC)algorithmwhichcanbeusedtoobtainsamplesfromtheposteriordistributionoftheparameters.Issuessuchasmodelt,modelselectionandinterpretationarediscussedinthecontextofadataanalysisoninternationalrelationsinSection5.AdiscussionfollowsinSection6.2LinearMixedEectsModelsforExchangeableDyadicDataSupposeweareonlyinterestedinestimatingthelinearrelationshipsbetweenresponsesyi;jandapossiblyvectorvaluedsetofvariablesxi;j,whichcouldincludecharacteristicsofuniti,character-isticsofunitj,orcharacteristicsspecictothepair.Inthiscasewemightconsidertheregressionmodelyi;j=0xi;j+i;j;(1)whereyi;iistypicallynotdened.Thegeneralizedleastsquaresestimate^anditscovariancematrixdependonthejointdistributionofthei;j'sonlythroughtheircovariance.Itisoftenassumedinregressionproblemsthattheregressorsxi;jcontainenoughinformationsothatthedistributionoftheerrorsisinvariantunderpermutationsoftheunitlabels.Thisassumptionisequivalenttothennmatrixoferrors(withanundeneddiagonal)havingadistributionthatis2 invariantunderidenticalrowandcolumnpermutations,sothatfi;j:i6=jgisequalindistributiontof(i);(j):i6=jgforanypermutationoff1;:::;ng.Thisconditioniscalledweakrow-and-columnexchangeabilityofanarray.Forundirecteddata,suchexchangeabilityimpliesa\randomeects"representationoftheerrors,inthati;jisequalindistributiontof(;ai;aj;\ri;j)where;ai;aj;\ri;jareindependentrandomvariablesandfisafunctiontobespecied(Aldous1985,Theorem14.11).IfinadditiontotheaboveinvarianceassumptionwealsomodeltheerrorsasGaussian,thenthejointdistributioncanberepresentedintermsofalinearrandomeectsmodel.Inthemoregeneralcaseofdirectedobservations,wecanrepresentthejointdistributionofthei;j'sasfollows:i;j=ai+bj+\ri;j(2)(ai;bi)0multivariatenormal(0;a;b);ab= 2aabab2b!(\ri;j;\rj;i)0multivariatenormal(0;\r);\r= 2\r2\r2\r2\r!;witheectsotherwisebeingindependent.Thecovariancestructureoftheerrors(andthustheobservations)isasfollows:E(2i;j)=2a+2ab+2b+2\rE(i;ji;k)=2aE(i;jj;i)=2\r+2abE(i;jk;j)=2bE(i;jk;l)=0E(i;jk;i)=abandso2arepresentsthedependenceofobservationshavingacommonsender,2bthatofobser-vationshavingacommonreceiver,andrepresentsthecorrelationofobservationswithinadyad(ofteninterpretedas\mutuality"or\reciprocity").Thishasbeencalledthe\socialrelations"or\roundrobin"model(Warneretal.1979;Wong1982),andisrelatedtoamodelfordiallelcrossdatausedbyCockerhamandWeir(1977).Themodelisaspecialcaseofalineargroupsymmetrymodel(AnderssonandMadsen,1998),andhasbeenstudiedinthiscontextbyLi(2002).RecentadvancesinvariancecomponentestimationhavebeenmadebyGillandSwartz(2001)andLiandLoken(2002).Toanalyzeresponsesinparticularsamplespaces,theerrorstructuredescribedabovecanbeaddedtoalinearpredictorinageneralizedlinearmodel:i;j=0xi;j+ai+bj+\ri;j(3)E(yi;jji;j)=g(i;j)p(y1;2:::;yn;n 1j1;2:::;n;n 1)=Yi6=jp(yi;jji;j):Thisisageneralizedlinearmixed-eectsmodelwithinverse-linkfunctiong(),inwhichtheobser-vationsaremodeledasconditionallyindependentgiventherandomeects,butareunconditionally3 dependent.ThecovariancepatternfortheobservationsisgivenapproximatelyasCov(yi1;j1;yi2;j2)=E[Cov(yi1;j1;yi2;j2ji1;j1;i2;j2)]+Cov[E(yi1;j1ji1;j1);E(yi2;j2ji2;j2)]=E[0]+Cov[g(i1;j1);g(i2;j2)]Cov(i1;j1;i2;j2)g0(0xi1;j1)g0(0xi2;j2);wherethepatternforCov(i1;j1;i2;j2)isthesameasthatforthei;j'sgivenabove.However,unlikethelinearregressioncase,^isnotgivenbylinearcombinationsoftheobservations,andE(^)andCov(^)arenotfunctionsofonlytherstandsecondordermomentsofthedata.Modellackoft,orthirdandhigherorderdependence,willaectourinferenceon.Manydyadicdatasetsexhibitcertainformsofthirdorderdependence.Indeed,itisthesehigherorderpatternsofdependencethatareoftenofinterest,andmayalsoprovideinformationusefulforpredictiveinference.3ModelingThirdOrderDependencePatternsSomedependencepatternscommonlyseenindydaicdatasetshavebeengiventhedescriptivetitlesoftransitivity,balance,andclusterability.Inthecontextofbinarydata,graphtheoreticdenitionsoftheseconceptsappearinWassermanandFaust(1994,chapter6)andareasfollows:Transitivity:Fordirectedbinarydata,anorderedtriadi;j;kistransitiveifwheneveryi;j=1andyj;k=1,wehaveyi;k=1,i.e.\afriendofafriendisafriend."Balance:Forsignedunorderedrelations,atriadi;j;kissaidtobebalancedifyi;jyj;kyk;i0.Theideaisthatiftherelationshipbetweeniandjis\positive"thentheywillrelatetoanotherunitkinanidenticalfashion,sothatifyi;j0thenyj;kandyk;iareeitherbothpositiveorbothnegative.Clusterability:Thisisarelaxationoftheconceptofbalance.Atriadisclusterableifitisbalancedortherelationsareallnegative.Theideaisthataclusterabletriadcanbedividedintogroupswherethemeasurementsarepositivewithingroupsandnegativebetweengroups.Inastatisticalsense,adatasetwilldisplayvaryingdegreesoftransitivity,balance,orclusterability.Oftenitisfoundthattherearemoretransitive,balanced,orclusterabletriadsthanwouldbeexpectedundermodels(2)or(3).Anotherindicationofthirdorderdependencewouldbeifafterttingaregressionmodelandobtainingtheresiduals^i;j,theaveragevalueof^i;j^j;k^k;iissubstantiallylargerthanzero,theexpectedvaluepresumedbymodel(2).Hoetal.(2002)usedsimplefunctionsoflatentcharacteristicvectorsinaxedeectssettingtocapturesomeformsoftransitivity,balance,andclusterability.Forexample,theyconsideredmodelsinwhichi;j=0xi;j+f(zi;zj)wheref(zi;zj)= jzi zjj(\thedistancemodel")orf(zi;zj)=z0izj=jzjj(\theprojectionmodel").Inwhatfollows,weconsiderasimilarapproachusingtheinnerproductkernelf(zi;zj)=z0izj,andgiverandomandxedeectsinterpretations.4 Addingthebilineareectz0izjtothelinearrandomeectsinmodels(2)and(3)givesi;j=ai+bj+\ri;j+i;j(4)i;j=z0izjwheretherandomeectsai;bjand\ri;jaremodeledwiththemultivariatenormaldistributionsdescribedabove.Wehavewritteni;j=z0izjtosuggesttheinterpretationofz0izjasamean-zerorandomeect:Ifthez'saremodeledasindependentk-dimensionalmultivariatenormalrandomvectorswithmeanzeroandcovariancematrixz,thentheresultingdistributionforthe'shasthefollowingmomentproperties:E(i;j)=0;E(2i;j)=trace2z;E(i;jj;kk;i)=trace3z;withallothersecondandthirdordermomentsequaltozero.Notethatanorthogonaltransfor-mationofthez'sleavesz0iziinvariant,sowecanassumezisadiagonalmatrix(otherwise,theo-diagonaltermsarenon-identiable).Forsimplicitywefocusonthecasez=2zIkk,forwhichtheabovemomentsare0;k4z,andk6zrespectively.Withi;jaddedtotheerrorterm,thenonzerosecondandthirdordermomentsareE(2i;j)=2a+2ab+2b+2\r+k4zE(i;ji;k)=2aE(i;jj;i)=2\r+2ab+k4zE(i;jk;j)=2bE(i;jj;kk;i)=k6zE(i;jk;i)=ab:Thustheeecti;j=z0izjcanbeinterpretedasamean-zerorandomeectabletoinduceaparticularformofthird-orderdependenceoftenfoundindyadicdatasets.Marginally,askincreasesthedistributionofi;jwillconvergetoanormaldistribution,duetothecentrallimittheorem.Jointly,theMarkovdependencegraphforthe'shastwodyadsasneighborsiftheyhaveatleastoneunitincommon.Consideredasxedeects,the'scanbeviewedasinteractiontermsthatarehighlyconstrainedduetothefunctionaldependenceonthez's.Theconstraintiseasytovisualizeintermsofthez's:Ifziandzjarevectorsofsimilardirectionandmagnitude,thenz0izkandz0jzkwillnotbetoodierent.Thisfeaturecanberelatedtotransitivity,whichisconceptuallyameasureofhowi;kisafunctionofi;jandj;k.Consideringforthemomentz'sscaledtohaveunitlengthsothatjzi zjj=p2(1 z0izj),bythetriangleinequalitywehave1 z0izk1 z0izj+1 z0jzk+2q(1 z0izj)(1 z0jzk);ori;ki;j+j;k 1+2q(1 i;j)(1 j;k);5 whichgivesalowerboundfori;kintermsofi;jandj;k.Balanceandclusterabilitydescribehowsimilari;kandj;kareasafunctionofi;j.Forscaledz's,wehaveji;k j;kj=jz0k(zi zj)jjzkjjzi zjj=jzi zjj:Notingthatz0izj=cos(i j),whereiistheangleofzifromaxedaxis,wehavejzi zjj=2sin[(i j)=2]=2sin12cos 1(z0izj)=q2(1 i;j);andsoji;k j;kjp2(1 i;j).Ifi;jislarge,thedierencebetweeni;kandj;kmustbesmall.Ifi;jisnegativeone,thedierenceisunconstrainedandcouldrangefromzerotoamaximumoftwo(inthisscaledcase).4ParameterEstimationInthefrequentistsetting,approximateestimationforgeneralizedlinearmixedeectsmodelsoftenproceedsviaTaylorexpansionsanditerativelyreweightedleastsquaresforthexedeects,alongwithapproximaterestrictedmaximumlikelihoodestimationforthevariancecomponents(Schall1991;BreslowandClayton1993;WolngerandO'Connell1993;McGilchrist1994).Theaccuracyoftheseapproximatemethodsisgenerallydependentonthesamplesize,seeBoothandHobert(1998)foradiscussion.Gabriel(1998)suggestsanalgorithmalongtheselinesforthegeneralizedbilinearmixedeectsmodel.Alternatively,ZegerandKarim(1991),Gelfand,SahuandCarlin(1996),andNatarajanandKass(2000)haveproposedGibbssamplingapproachestoparameterestimationforgeneralizedlinearmixedeectsmodels.However,estimationismoredicultforthecomplicateddependencestructureoftherandomeectsintheinvariantnormalmodel(2).GillandSwartz(2001)haveproposedaGibbssamplingschemeforestimationofrandomeectsinthelinearcasewiththeidentitylink,althoughwehavefoundthattheiralgorithmdoesnotmixwellwhencovariatesareincluded,duetoaweakidentiabilityoftheunitlevelrandomeectsandcertainregressioncoecients:AsdiscussedinGelfand,Sahu,andCarlin(1995)therandomeectsaandbwillbeconfoundedtoadegreewitheachotherandtoregressionparametersassociatedwithpredictorsthatdonotvaryacrossreceivers(i.e.sender-speciceects)oracrosssenders(receiver-speciceects).Forexample,apopulation-levelinterceptisonesuchparameter.Toobtaina\cleaner"partitionofthevarianceandamoreecientMCMCsamplingscheme,wedecomposexi;jintoxi;j=(xd;i;j;xs;i;xr;j),i.e.intodyadspecicregressorsxd;i;j,senderspecicregressorsxs;iandreceiverspecicregressorsxr;j.Thegeneralizedbilinearmodelisthenrewrittenasi;j=0dxd;i;j+(0sxs;i+ai)+(0rxr;j+bj)+\ri;j+z0izj6 orequivalentlyi;j=0dxd;i;j+si+rj+\ri;j+z0izjsi=0sxs;i+airi=0rxr;i+bi:Thisparameterizationforthelinearunit-leveleectsissimilartothe\centered"parameterizationssuggestedbyGelfandetal.(1995,1996).Notethataninterceptcanbethoughtofasbothasenderorreceiverspeciceect.Forsymmetry,weincludetheconstant1/2atthebeginningofeachxs;iandxr;jvector,andestimatetherstcomponentsofsandrasbeingequal.Usingtheabovereparameterizationfori;j,weestimatetheparametersforthegeneralizedbilinearregressionmodelbyconstructingaMarkovchaininfd;s;r;ab;Z;2z;\rg(whereZdenotestheknmatrixoflatentvectors),havingp(d;s;r;ab;Z;2z;\rjY)astheinvariantdistribution.ThisisobtainedviaanalgorithmbasedonGibbssampling,whichalsosampless;randthe's.Thebasicalgorithmistoiteratethefollowingsteps:1.Samplelineareects:(a)Sampled;s;rjs;r;ab;\r;;Z(linearregression);(b)Samples;rjs;r;ab(linearregression);(c)Sampleaband\rfromtheirfullconditionals.2.Samplebilineareects:(a)Fori=1;:::;n:samplezijfzj;j6=ig;;;s;r;z;\r(alinearregression);(b)Samplezfromitsfullconditional.3.Sampledyadspecicparameters:Updatefi;j;j;igusingaMetropolis-Hastingsstep:(a)Propose(i;jj;i)MVN((0xi;j+ai+bj+z0izj0xj;i+aj+bi+z0jzi);\r);(b)Accept(i;jj;i)withprobabilityp(yi;jji;j)p(yj;ijj;i)p(yi;jji;j)p(yj;ijj;i)^1:Variouscombinationsoftheabovestepscanbeusedtoestimatedierentmodels.Thestepsin1aloneprovideaBayesianestimationprocedureforthelinearregressionproblemhavinganerrorcovarianceasin(2).Bayesianestimationofthenormalbilinearmodelwiththeidentitylinkcouldproceedbyreplacingeachi;jwithyi;jandonlyiteratingsteps1and2.Estimationofageneralizedlinearmixedeectsmodelwithrandomeectsstructuregivenby(2)couldproceedbyiteratingsteps1and3.Thefullconditionaldistributionsrequiredtoperformsteps1and2aregivenbelow.Notethatthe'sareessentiallyunrestrictedintheabovesamplingscheme.Atthislevelthetissaturatedanddoesnotdependontheregressors,atleasttothedegreethatthepriorfor7 \risdiuse.WhattheMCMCalgorithmaboveprovidesisessentiallyasaturatedtforthe's(althoughsomewhatsmoothedbythecommonvariance)andanANOVA-likedecompositionofthe'sintoregressor,sender,receiverandinner-producteects.4.1ConditionalDistributionsfortheLinearEectsComponents:Notingthati;j z0izj=0dxi;j+si+rj+\ri;j,weseethatconditionalonthe'sandz's,theotherparameterscanbesampledusingastandardBayesiannormal-theoryregressionapproach,althoughwithacomplicatedcovariancestructure.Fullconditionalof(d;s;r):SimilartoWong's(1982)approachtotheinvariantnormalmodel,weletui;j=i;j+j;i 2z0izjandvi;j=i;j j;iforij.Wethenhave uv!= XuXv!0BB@dsr1CCA+ uv!;(5)whereXuandXvaretheappropriatedesignmatricesanduandvarevectorsofindependenterrortermswithvariances2u=22\r(1+)and2v=22\r(1 )respectively.Thefullconditionaldistributionof(d;s;r)isthenproportionaltop(u;vjd;s;r;\r)p(s;rjs;r;ab)p(d).Foramultivariatenormal(d;d)priordistributionond,thetermintheexponentofthefullconditionalis0" 1dd 1srXsrsr!+X0uu=2u+X0vv=2v# 120" 1d00 1sr!+X0uXu=2u+X0vXv=2v#where0=(0ds0r0),Xsrandsrarethecombineddesignmatrixandregressionparametersforsandr,andsristhecovariancematrixof(s0r0)0,whichiseasilyderivedfromab.Theconditionaldistributionisthusmultivariatenormal(;)where=" 1dd0 1srXsrsr!+X0uu=2u+X0vv=2v#=" 1d00 1sr!+X0uXu=2u+X0vXv=2v# 1:Notethattheinverseofsrisgivenby 1sr= (2b=)Inn (ab=)Inn (ab=)Inn(2a=)Inn!;=2a2b 2ab:Fullconditionalof(s;r):Thefullconditionalof(s;r)isproportionaltop(s;rjs;r;ab)p(s;r).Assumingamultivariatenormal(sr;sr)priordistributionforthecombinedregression8 parameters,thefullconditionalisamultivariatenormaldistributionwithmeanandvariance(;)givenby=" 1srsr+Xsr 1sr sr!#=( 1s;r+X0sr 1srXsr) 1Fullconditionalofab:Thefullconditionalofabisproportionaltop(s;rjs;r;ab)p(ab).UsingapriordistributionofabinverseWishart(ab0;)(parameterizedsothatE(ab)=ab0=( 3)),thefullconditionalofabisabja;binverseWishart(ab0+(ab)0(ab);+n),wherea=(s Xss)andb=(r Xrr).Fullconditionalof\r:Usingpriordistributionsof2uinversegamma(u1;u2)and2vinversegamma(v1;v2),thefullconditionalsaregivenby2ujuinversegamma(u1+12 n2;u2+12P[ui ^ui;j]2)and2vjvinversegamma(v1+12 n2;v2+12P[vi ^vi;j]2),where^ui;j=E[ui;jjd;xi;j;si;rj]=0d(xi;j+xj;i)+si+sj+ri+rj,and^vi;jisgivensimilarly.Thecovariancematrix\rcanbere-constructedfrom2uand2vvia2\r=(2u+2v)=4and=(2u 2v)=(2u+2v).4.2ConditionaldistributionsfortheBilinearEectsComponent:Letei;j=(i;j+j;i ^ui;j)=2,theresidualofthesymmetricpartofthematrixof'safterttingthelineareects,andletu;i;j=\ri;j+\rj;i.Consideringthefullconditionalofzi,wehaveei;1=z0iz1+u;i;1=2ei;2=z0iz2+u;i;2=2...ei;n=z0izn+u;i;n=2;andweseethatsamplingzifromitsfullconditionalisequivalenttoa(Bayesian)linearregressionproblem.Modelingthez'sasaprioriindependentmultivariatenormal(0;z)variables,thefullconditionalofziismultivariatenormal(;)with=4Z iei; i=2u=( 1z+4Z iZ0 i=2u) 1whereZ idenotesthek(n 1)matrixobtainedbyremovingtheithcolumnofZ,andei; idenotesthevectorofresidualsfei;j:j6=ig.Usinganinverse-Wishart(z0;)prior,thefullconditionalofzisinverse-Wishart(z0+ZZ0;+n).Alternatively,ifwerestrictztobe2zIkkanduseaninversegamma(0;1)prior,thenthefullconditionalisgivenby2zjZinversegamma(0+(nk)=2;1+trace(Z0Z)=2).9 5DataAnalysis:InternationalRelationsinCentralAsiaWeanalyzedataoninternationalrelationsincentralAsiaasrecordedbytheKansasEventDataProject(http://www.ku.edu/keds/project.html)anddescribedbySchrodt,Simpson,andGerner(2001).NewsstoriesaredownloadedfromtheReutersBusinessBriengServiceonAfghanistan,Armenia,Azerbaijan,andtheformerSovietRepublicsofCentralAsia,andpoliticalinteractionsbetweencountriesarerecordedandcategorized.Wetakeourresponseyi;jtobethetotalnumberof\positive"actionsreportedlyinitiatedbycountryiwithtargetjfrom1992to1999(i.e.afterthebreakupoftheSovietUnion),asrecordedbytheKEDSproject.Positiveactionshereincludesucheventsasapproval,endorsementorpraiseofonegovernmentbyanother,militaryassistance,formationofalliances,promisesofnancialorpolicysupportandothers(essentiallyalleventshavingGoldsteinscalevaluesgreaterthan2.5,exceptcease-reorcedingofpower.SeetheKEDSprojectwebpageformoredetails).Weincludeinourpopulationthe99countriesclosestingeographicdistancetoAfghanistan,plustheUnitedStates,givingatotalofn=100countriesforanalysis.Wenotethatseventeenoftheone-hundredcountrieshadzeroactionsaseitherinitiatorsortargetsofactionsoverthesevenyearperiod.5.1DataDescriptionSomedescriptiveplotsoftherawdataaregiveninFigure1.Panel(a)plotslog(1+Pj:j6=iyi;j)versuslog(1+Pj:j6=iyj;i)foreachcountryi.ThequantitiesPj:j6=iyi;jandPj:j6=iyj;iaretypicallycalledtheoutdegreeandindegreeofuniti,respectively.Notethestrongcorrelation,whichsuggestsalargevalueofab=(ab)intherandomeectsmodelbeingconsidered.Inpanel(b)weplotthelogofeachcountry'soutdegreeplusone,log(1+Pj:j6=iyi;j),versuslogpopulation,whichsuggestsapositiverelationshipbetweenresponseandpopulation(aplotoflog-indegreeversuspopulationissimilar).Inpanel(c)weplottheresponseonalogscaleversusthegeographicdistanceinthousandsofmilesbetweencountriesiandj.Moreprecisely,thisdistanceisthe\minimumdistance"betweentwocountries,andiszeroifiandjshareaborder.Onaverage,thenumberofeventsbetweentwocountriesdecreasesasgeographicdistanceincreases.ThispatternismademoreclearbyseparatingoutthemeasurementsinvolvingtheUnitedStates(whicharecircled).5.2ModelandPriorsNotethatthedataarefromanobservationalstudy,andthatthedataarenotrandomlysampled.Rather,wehavedenedapopulationofunitsbasedongeographicdistanceandhavemeasurementsonallpairs.Forthisanalysis,weprimarilyinterpretaprobabilitymodelasatoolfordescribingthevarianceinthedataset,andtheregressioncoecientsasmeasuresofthemultiplicative,orlog-linear,componentsoftherelationshipbetweenresponseandregressors.Wettherandomeectsmodel(4)tothedatausingaPoissondistributionandthelog-link,10 012345670123456log(indegree+1)log(outdegree+1)AfghanistanAlbaniaAlgeriaArmeniaAustriaAzerbaijanBahrainBangladeshBelarusBelgiumBhutanBosnia and HerzegovinaBrunei DarussalamBulgariaBurundiCambodiaComorosCroatiaCyprusCzech RepublicDenmarkDjiboutiEgyptEstoniaEthiopiaFinlandFranceGeorgiaGermanyGreeceHong KongHungaryIndiaIndonesiaIranIraqIsraelItalyJordanKazakstanKenyaKoreaKuwaitKyrgyzstanLaosLatviaLebanonLibyaLithuaniaMacauMacedoniaMalaysiaMaldivesMaltaMauritaniaMoldovaMongoliaMyanmarNepalNetherlandsNorth KoreaNorwayOmanPakistanPalestinePhilippinesPolandPR ChinaQatarRomaniaRussiaRwandaSaudi ArabiaSenegalSerbiaSeychellesSingaporeSlovakiaSloveniaSomaliaSri LankaSudanSwedenSwitzerlandSyrian Arab RepublicTaiwanTajikistanTanzaniaThailandTunisiaTurkeyTurkmenistanUgandaUkraineUnited Arab EmiratesUnited KingdomUnited StatesUzbekistanViet NamYemen(a)1012141618200123456log(population)log(outdegree+1)AfghanistanAlbaniaAlgeriaArmeniaAustriaAzerbaijanBahrainBangladeshBelarusBelgiumBhutanBosnia and HerzegovinaBrunei DarussalamBulgariaBurundiCambodiaComorosCroatiaCyprusCzech RepublicDenmarkDjiboutiEgyptEstoniaEthiopiaFinlandFranceGeorgiaGermanyGreeceHong KongHungaryIndiaIndonesiaIranIraqIsraelItalyJordanKazakstanKenyaKoreaKuwaitKyrgyzstanLaosLatviaLebanonLibyaLithuaniaMacauMacedoniaMalaysiaMaldivesMaltaMauritaniaMoldovaMongoliaMyanmarNepalNetherlandsNorth KoreaNorwayOmanPakistanPalestinePhilippinesPolandPR ChinaQatarRomaniaRussiaRwandaSaudi ArabiaSenegalSerbiaSeychellesSingaporeSlovakiaSloveniaSomaliaSri LankaSudanSwedenSwitzerlandSyrian Arab RepublicTaiwanTajikistanTanzaniaThailandTunisiaTurkeyTurkmenistanUgandaUkraineUnited Arab EmiratesUnited KingdomUnited StatesUzbekistanViet NamYemen(b)0510150123456geographic distancelog(yij+1)(c)Figure1:Relationshipsbetween(a)Outdegreeandindegree;(b)Outdegreeandpopulation;(c)Responseandgeographicdistance.ResponsesinvolvingtheUnitedStatesarecircled.sothateachresponseyi;jisassumedtohavecomefromaPoissondistributionwithmeanei;j,andthatthey'sareconditionallyindependentgiventhe's.Wedecomposethevarianceinthe'sasfollows:i;j=0+dxi;j+sxi+rxj+i;ji;j=ai+bj+\ri;j+z0izj;wherexi;jisthegeographicdistancebetweeniandjandxiisthelogpopulationofi.Forestimationofvariancecomponents,wemodeltherandomeectsashavingthefollowingmultivariatenormaldistributions:(ai;bi)0MVN(0;ab),(\ri;j;\rj;i)0MVN(0;\r),ziMVN(0;2zIkk).Priordistributionsoftheparametersaretakentobemultivariatenormal(0;100I44);abinverseWishart(I22;4);2u;2vi.i.d.inversegamma(1;1),2\r=(2u+2v)=4,=(2u 2v)=(2u+2v).PosteriorcalculationsproceedasdescribedinSection4.5.3SelectingtheLatentDimension:Oneissueinmodelttingistheselectionofthedimensionkofthelatentvariablesz.Selectionofkcoulddependonthegoaloftheanalysis.Forexample,ifthegoalisdescriptive,i.e.thedesiredendresultisadecompositionofthevarianceintointerpretablecomponents,thenachoiceofk=1;2or3wouldallowforasimplegraphicalpresentationofamultiplicativecomponentofthevariance.11 kLLP(k)logp(yj^;^a;^b;^Z;^)AIC^20-3558.78-2432.67-2638.672.381-3351.76-2317.47-2623.471.662-3078.79-2214.68-2620.681.233-3076.73-2127.26-2633.260.874-3077.30-2038.95-2644.950.54Table1:SelectionofkAlternatively,onecouldexaminemodeltasafunctionofkbasedonthelog-likelihood,oruseacross-validationcriterionifoneisprimarilyconcernedwithpredictiveperformance.Consideringlikelihood-basedmeasuresoft,thelog-probabilityofthedatagiventhevaluesoftheparametersgetsevaluatedforeachupdateofthe's,andsologp(Yj)=Pi6=jlogp(yi;jji;j)canbecalculatedwithnoextraeort.However,suchaquantityisnotappropriateforselectingbetweenmodels.AsdescribedinSection4,themodelisessentiallyunrestrictedinthe's,givinganearlysaturatedtwhichdoesnotdependmuchonthechoiceofkortheregressors(providedthepriorfor\rissucientlydiuse).Alikelihoodthatismoreappropriateisthemarginalprobabilityofdatawithinapair,logp(Yj;a;b;Z;\r)=P(i;j)logp(yi;j;yj;ij;ai;bj;aj;bi;zi;zj;\r),wherethesumisoverunorderedpairs.Thisisessentiallythelog-likelihoodtreatingthea;b,andz'sasxedeects.Notethatingenerallogp(yi;j;yj;ij;ai;bj;aj;bi;zi;zj;\r)isanintegralover\ri;jand\rj;ithatneedstobeapproximated,exceptinthecaseofthenormalmodelwiththeidentitylink.Insomesituationsthepurposeofthemodelistomakepredictionsofunobserveddata.Forexample,supposeonlyasubsetofthen(n 1)responseswererandomlychosentobemeasured.Aslongaswehavesomemeasurementsforeachunit,wecanestimatetheeectsa;bandzforeachunitandmakepredictionsformissingresponsesbasedontheseestimates.Althoughpredictionisnotthegoalforthesedata,forillustrativepurposeswecomparethemarginalprobabilitycriteriondiscussedabovetothefollowingfour-foldcrossvalidationprocedure:1.Randomlysplitthesetoforderedpairsfi;j:i6=jgintofourtestsetsA1;A2;A3;A4.2.Fork=0;1;2;3;4:(a)Forl=1;2;3;4:i.performtheMCMCalgorithmusingonlyfyi;j:fi;jg62Alg,butsamplevaluesofi;jforallorderedpairs.ii.Basedonthesampledvaluesofi;jcomputetheposteriormean^i;jforfi;jg2Alandthelogpredictiveprobabilitylpp(Al)=Pfi;jg2Allogp(yi;jj^i;j).(b)MeasurethepredictiveperformanceforkasLPP(k)=P4l=1lpp(Al).12 3.SelectkbasedonLPP(k).Forthesedata,themarginallikelihoodandcross-validationcriteriaforselectingkaregiveninTable2.Thecrossvalidationproceduresuggeststhatmodelshavingadimensionofk=2;3or4haveroughlythesamepredictiveperformance.Intermsofthemarginallikelihoodcriterion,thebiggestimprovementsintareingoingfromk=0tok=1andfromk=1tok=2.Theimprovementsintingoingfrom2to3andfrom3to4dimensionsaresmaller.UsinganAIC-likecriterionandpenalizingtheimprovementinlikelihoodbythenumberofadditionalparameters(100peradditionaldimension),wewouldchoosek=2.Basedontheseresults(andourabilitytoplotresultsintwo-dimensions)wechoosetopresenttheresultsforthek=2modelinmoredetail.050000100000150000200000-11-9-7-5iterationb0050000100000150000200000-0.30-0.20-0.10iterationbd0500001000001500002000000.40.81.21.6iterationbs0500001000001500002000000.40.81.21.6iterationbrFigure2:MarginalMCMCoutputforregressioncoecients.SolidlinesarefromtheMarkovchainwithdata-informedstartingvalues,dashedlinesfromthechainwithuninformedstartingvalues.5.4Resultsfork=2TwoMarkovchainsoflength200,000eachwereconstructedusingthealgorithmdescribedabove.Therstchainusedstartingvaluesofzeroforallregressioncoecientsandcountry-specicin-tercepts,theidentitymatrixforaband\r,avalueof0:1for2z,andcomponentsofZsampledindependentlyfromanormal(0;2z)distribution.Thesecondchainusedstartingvaluesobtainedfromthefollowingprocedure:Maximumlikelihoodestimatesofd,sandrwereobtainedbyt-tinganordinarygeneralizedlinearmodelusinggeographicdistanceasaregressorandsenderandreceiverlabelsasfactorvariables.Estimatesof0;s;r,andabwereobtainedfromtheesti-matesofsandr.Theiterativelyreweightedleast-squaresttingprocedureproducesamatrixRofworkingresiduals,withtheodiagonalelementsundened.Anestimate^ZofZwasthenobtained13 0500001000001500002000000246810iterationsa20500001000001500002000000246810iterationsb20500001000001500002000001.01.21.41.6iterationse20500001000001500002000001.01.52.02.53.0iterationsz2Figure3:MarginalMCMCoutputforvariancecomponentparameters.dsr2a2bab2\r2zmean-0.181.000.946.466.376.41.230.951.99sd0.040.170.171.231.21.210.140.010.27Table2:Posteriormeansandstandarddeviationsfork=2byapproximatingRwithamatrixproductoftheformZ0Z.Thiscanbedonewithaniterativeleast-squaresprocedure,similartotheGibbssamplingprocedureoutlinedinSection4.2:seetenBergeandKiers(1989)formoredetailsonthisproblem.Anestimateof\risthenobtainedfromR ^Z0^Z.SamplesofparametervaluesweresavedfromtheMarkovchainsevery100iterations,andareplottedinFigures2and3.Bothchainsappeartohaveachievedstationarityafterabout50,000iterations,andsowebaseourinferenceonthesavedsamplesafterthispoint.Posteriormeansandstandarddeviationsofthemodelparameters,basedonthe3000savedMCMCsamples(1500fromeachchain),aregiveninTable2.Asintherawdata,weseeanegativerelationbetweenresponseandgeographicdistance(E[djy]= 0:18),andapositiverelationbetweenresponseandcountrypopulations(E[sjy]=1:00;E[rjy]=0:94).Wealsoestimateastrongpositivecorrelationofwithin-dyadresponsesaswellasthewithin-countryrandomeectsaandb.Next,weanalyzetheposteriordistributionofthetheknmatrixoflatentvectorsZ.NotethattheprobabilitymodeldependsonZonlythroughthematrixofinnerproductsZ0Z,whichisinvariantunderrotationsandre\rectionsofZ.Therefore,logPr(YjZ;;X)=logPr(YjZ;;X)foranyZwhichisequivalenttoZundertheoperationsofrotationorre\rection.ValuesofZ14 -4-202-3-2-1012AfghanistanAlbaniaAlgeriaArmeniaAustriaAzerbaijanBahrainBangladeshBelarusBelgiumBhutanBosnia and HerzegovinaBrunei DarussalamBulgariaBurundiCambodiaComorosCroatiaCyprusCzech RepublicDenmarkDjiboutiEgyptEstoniaEthiopiaFinlandFranceGeorgiaGermanyGreeceHong KongHungaryIndiaIndonesiaIranIraqIsraelItalyJordanKazakstanKenyaKoreaKuwaitKyrgyzstanLaosLatviaLebanonLibyaLithuaniaMacauMacedoniaMalaysiaMaldivesMaltaMauritaniaMoldovaMongoliaMyanmarNepalNetherlandsNorth KoreaNorwayOmanPakistanPalestinePhilippinesPolandPR ChinaQatarRomaniaRussiaRwandaSaudi ArabiaSenegalSerbiaSeychellesSingaporeSlovakiaSloveniaSomaliaSri LankaSudanSwedenSwitzerlandSyrian Arab RepublicTaiwanTajikistanTanzaniaThailandTunisiaTurkeyTurkmenistanUgandaUkraineUnited Arab EmiratesUnited KingdomUnited StatesUzbekistanViet NamYemenFigure4:PosteriormeanofZsampledfromtheposteriordistributionmayseematrsttobehighlyvariable,butperhapsarenearlyrotationsofeachotherandarethusnothighlyvariableintermsoftheresultinginnerproductmatrices.ToappropriatelycomparesamplevaluesofZ,wemustrstrotatethemtoacommonorientation.Forthesedatathisisdoneusinga\Procrustean"transformation(Sibson1978),inwhichforeachsampleZwendtherotationZofZthathasthesmallestsumofsquareddeviationsfromanarbitraryxedreferencematrixZ0.TherotatedmatrixZwhichminimizesthesumofsquaresisgivenbyZ=Z0Z0(ZZ00Z0Z0) 1=2Z.SeeHoetal.(2002)forfurtherdiscussion.TheresultingmeanofZisgiveninFigure4.Marginaluncertaintyinthez'scouldbedisplayedbyplottingsamplez'sovertheplotofthemeans,usingcolorstodistinguishbetweencountries.Generally,twocountrieswillbemodeledashavingz'sinthesamedirectioniftheyhavelargeresponsestooneanotherrelativetotheirtotalnumberofactionsandcovariatevalues,and/oriftheirresponsesinvolvingothercountriesaresimilar(amodelwhichcandistinguishbetweenthesetwophenomenaisproposedinthediscussion).Forexample,CroatiaandSloveniaareeachrecordedastheinitiatorofanactionwiththeotherasatarget,andeachinitiatesanactionwithSerbiaaswell.WiththeexceptionofoneactionfromSloveniatoItaly,thesearetheonlyeventsrecordedforCroatiaandSlovenia,andsothesecountriesare\similar"inthattheyhaveactionsinvolvingeachotherandtoSerbia,andonlyoneotheractioninvolvinganothercountry.Bosnia-HerzegovinaandDenmarkhavenoactionswithCroatiaorSlovenia,butlikeCroatiaandSloveniatheyeach15 haveoneactionwithSerbiaandveryfewactionsotherwise(eachhasoneactionwithAzerbaijan,andnootheractions),andarethuslocatedinasimilardirection.Serbia,althoughactivewiththisgroupofcountries(onthescaleoftheirresponserates),hasactionswith10othercountries,andisthereforeplacedmoretowardsthecenter.Ofcourse,theposteriorvariancesofthez'sforCroatia,Slovenia,Bosnia-Herzegovina,andDenmarkarequitehigh,asourinformationaboutthemiscomingprimarilyfromthefewnonzeroresponsesamongthem.var[log(yij+1)]Density0.180.200.220.240.260.2805101520(a)0.00.51.01.50.00.51.01.52.0sender specific variancepredicted sender specific variance(b)Figure5:Goodnessofttests:(a)Posteriorpredictivedistributionofpopulationvariance.(b)Posteriorpredictivecondenceregionsforcountry-specicvarianceinactioninitiation.Finally,weevaluatesomeaspectsofmodeladequacyviagoodnessoftstatistics.ThisisdonebycomparingtheobservedvalueofastatisticofinterestT(Y)toitsposteriorpredictivedistributionp(T(Ypred)jY).SamplesfromtheposteriorpredictivedistributionareobtainedbysimulatingdatasetsusingtheparameterssampledbytheMarkovchain(see,forexampleGelman,Carlin,SternandRubin1995chapter6).InthepresentcasewemightbeinterestedinanyoverorunderdispersionofthedatarelativetothePoissonmodel.Weevaluateanysuchlackoftbyconsideringasteststatisticstheoverallsamplevarianceoflog(yi;j+1),aswellasthesamplevarianceofflog(yi;j+1):j6=igforeachi,thatis,thevarianceofresponsesfromeachsender,onalogscale.Theposteriorpredictivedistributionsofthesequantitieswereestimatedbysub-sampling1000valuesof(d;s;r;Z;\r)fromthetwoMarkovchains,generatingadatasetfromeachsub-sampledsetofparametervalues,andthencomputingthestatisticsfromeachgenerateddataset.TheresultsareplottedinFigure5.Therstpanelgivesahistogramof1000samplesfromtheposteriorpredictivedistributionoftheoverallvariance.Theposteriorpredictivedistributioniscenteredaroundtheobservedoverallvariance,givenbytheverticalline,andnolackoftisindicatedbythisstatistic.ThesecondpanelofFigure5plotstheobservedsender-specicvariancesforeachcountryversusa95%posteriorpredictive16 intervalforthatquantity.Thecondenceintervalscontaintheobservedvaluesfor97ofthe100countries,andthusdonotindicatemuchlack-of-t.ThePoissonmodelseemstotthevarianceinresponsereasonablywell,atleastintermsofthesestatistics.6DiscussionThisarticlehaspresentedanapproachtomodelingthirdorderdependencepatternsoftenseenindyadicdatasets,suchassocialnetworks.Themodelsarebasedongeneralizedlinearmixedeectsmodelswiththeadditionofareduced-rankinteractiontermcomposedofinnerproductsoflatentcharacteristicvectors.Suchanapproachallowsfortheanalysisofdyadicdatausingfamiliarregressiontools,butalsoallowsonetocapturepatternssuchastransitivity,balance,andclusterabilitywhichareoftenofinteresttosocialscienceresearchers.Otherapproachestocapturingsuchdependencepatternshaveusedmetricdistances(Hoetal.2002)andultrametricdistances(SchweinbergerandSnijders,2003),althoughnotinthepresenceofthecovariancestructure(2).Whilesuchlatentdistancemodelsmaybeeasytounderstand,theinner-productapproachhassomeconceptualappeal,asthetermz0izjcanbeviewedasamean-zerorandomeect.Anotherdependencepatternoftenofinteresttoresearchersisthatofstochasticequivalence,inwhichtwounitsiandjaresaidtobestochasticallyequivalentiftheirresponseshavethesameprobabilitydistribution,i.e.p(yi;1;:::;yi;n)=p(yj;1;:::;yj;n).Themodelconsideredinthispaper,aswellasthelatentdistanceapproachesmentionedabove,potentiallyconfoundstochasticequivalencepatternswiththoseofclusterabilityandbalance:twounitswillgenerallybeestimatedtohavesimilarlatentcharacteristicvectorsiftheyhavestrongrelationstoeachother,orhavesimilarrelationstoothersunitunits.However,insomedatasetstheremaybeclustersofunitsthatrelatesimilarlytoothers,butnotstronglytoeachother.NowickiandSnijders(2001)consideredalatentclassmodelwhichidentiedclustersofsuchstochasticallyequivalentunits,butdidnotseparatelyconsiderclusteringbasedonstrengthofrelations.Apossibleapproachtomodelingbothtypesofpatternsistoextendthebilineareectdiscussedinthispapertoamoregeneralasymmetricbilineareectsuchasz0iRzj,whereRisakkmatrix.EstimationofsimilartypesofeectshasbeenconsideredbybyGabriel(1998),andleastsquaresrepresentationsofanasymmetricmatrixYbyZ0RZhasbeenconsideredbytenBergeandKiers(1989),Kiers(1989)andTrendalov(2002),amongothers.Inthepresentapplication,thevectorzicouldbeinterpretedasgivinggradesofmembershipforunititoeachofkclasses,andRlmastheresponseratefromclassltom.Interestingly,therestrictionofeachzitobeunityatonecomponentandzeroattheothersgivesarepresentationofthelatentclassmodelofNowickiandSnijders(2000).Unrestrictedestimationofz0iRzj,inthepresenceoftheerrorstructure(2),isatopicofcurrentresearchbytheauthor.17 ReferencesAldous,D.J.(1985),\Exchangeabilityandrelatedtopics,"inEcoled'etedeprobabilitesdeSaint-Flour,XIII|1983,vol.1117ofLectureNotesinMath.,pp.1{198,Springer,Berlin.Andersson,S.andMadsen,J.(1998),\Symmetryandlatticeconditionalindependenceinamulti-variatenormaldistribution,"TheAnnalsofStatistics,26,525{572.Booth,J.G.andHobert,J.P.(1998),\Standarderrorsofpredictioningeneralizedlinearmixedmodels,"JournaloftheAmericanStatisticalAssociation,93,262{272.Breslow,N.E.andClayton,D.G.(1993),\Approximateinferenceingeneralizedlinearmixedmodels,"JournaloftheAmericanStatisticalAssociation,88,9{25.Cockerham,C.C.andWeir,B.S.(1977),\Quadraticanalysesofreciprocalcrosses,"Biometrics,33,187{204.Gabriel,K.R.(1978),\Leastsquaresapproximationofmatricesbyadditiveandmultiplicativemodels,"JournaloftheRoyalStatisticalSociety.SeriesB.Methodological,40,186{196.Gabriel,K.R.(1998),\Generalisedbilinearregression,"Biometrika,85,689{700.Gelfand,A.E.,Sahu,S.K.,andCarlin,B.P.(1995),\Ecientparameterisationsfornormallinearmixedmodels,"Biometrika,82,479{488.Gelfand,A.E.,Sahu,S.K.,andCarlin,B.P.(1996),\Ecientparametrizationsforgeneralizedlinearmixedmodels,"inBayesianstatistics,5(Alicante,1994),OxfordSci.Publ.,pp.165{180,OxfordUniv.Press,NewYork.Gelman,A.,Carlin,J.B.,Stern,H.S.,andRubin,D.B.(1995),Bayesiandataanalysis,Chapman&HallLtd,London.Gill,P.S.andSwartz,T.B.(2001),\Statisticalanalysesforroundrobininteractiondata,"TheCanadianJournalofStatistics,29,321{331.Ho,P.D.,Raftery,A.E.,andHandcock,M.S.(2002),\Latentspaceapproachestosocialnetworkanalysis,"JournaloftheAmericanStatisticalAssociation,97,1090{1098.Kiers,H.A.L.(1989),\Analternatingleastsquaresalgorithmforttingthetwo-andthree-wayDEDICOMmodelandtheIDIOSCALmodel,"Psychometrika,54,515{521.Li,H.(2002),\Modelingthroughgroupinvariance:aninterestingexamplewithpotentialapplica-tions,"TheAnnalsofStatistics,30,1069{1080.Li,H.andLoken,E.(2002),\Auniedtheoryofstatisticalanalysisandinferenceforvariancecomponentmodelsfordyadicdata,"StatisticaSinica,12,519{535.18 Marasinghe,M.G.andJohnson,D.E.(1982),\Atestofincompleteadditivityinthemultiplicativeinteractionmodel,"JournaloftheAmericanStatisticalAssociation,77,869{877.McGilchrist,C.A.(1994),\Estimationingeneralizedmixedmodels,"JournaloftheRoyalStatis-ticalSociety.SeriesB.Methodological,56,61{69.Natarajan,R.andKass,R.E.(2000),\ReferenceBayesianmethodsforgeneralizedlinearmixedmodels,"JournaloftheAmericanStatisticalAssociation,95,227{237.Nowicki,K.andSnijders,T.A.B.(2001),\Estimationandpredictionforstochasticblockstruc-tures,"JournaloftheAmericanStatisticalAssociation,96,1077{1087.Oman,S.D.(1991),\Multiplicativeeectsinmixedmodelanalysisofvariance,"Biometrika,78,729{739.Schall,R.(1991),\Estimationingeneralizedlinearmodelswithrandomeects,"Biometrika,78,719{727.Schrodt,P.A.,Simpson,E.M.,andGerner,D.J.(2001),\Monitoringcon\rictusingautomatedcod-ingofnewswiresources,"PaperpresentedattheHigh-LevelScienticConferenceonIdentifyingWars,UppsalaUniversity,Uppsala,Sweden.http://www.pcr.uu.se/Schrodt_Uppsala.pdf.Schweinberger,M.andSnijders,T.(2003),\Settingsinsocialnetworks:Ameasurementmodel."Submitted.Sibson,R.(1978),\Studiesintherobustnessofmultidimensionalscaling,"JournaloftheRoyalStatisticalSociety,SeriesB,Methodological,40,234{238.tenBerge,J.M.F.andKiers,H.A.L.(1989),\Fittingtheo-diagonalDEDICOMmodelintheleast-squaressensebyageneralizationoftheHarmanandJonesMINRESprocedureoffactoranalysis,"Psychometrika,54,333{337.Trendalov,N.T.(2002),\GIPSCALrevisited.Aprojectedgradientapproach,"StatisticsandComputing,12,135{145.Warner,R.,Kenny,D.A.,andStoto,M.(1979),\Anewroundrobinanalysisofvarianceforsocialinteractiondata,"JournalofPersonalityandSocialPsychology,37,1742{1757.Wasserman,S.andFaust,K.(1994),SocialNetworkAnalysis:MethodsandApplications,Cam-bridgeUniversityPress,Cambridge.Wasserman,S.andPattison,P.(1996),\Logitmodelsandlogisticregressionsforsocialnetworks:I.AnintroductiontoMarkovgraphsandp*,"Psychometrika,61,401{425.19 Wolnger,R.andO'Connell,M.(1993),\Generalizedlinearmixedmodels:Apseudo-likelihoodapproach,"JournalofStatisticalComputationandSimulation,48,233{243.Wong,G.Y.(1982),\Roundrobinanalysisofvarianceviamaximumlikelihood,"JournaloftheAmericanStatisticalAssociation,77,714{724.Zeger,S.L.andKarim,M.R.(1991),\Generalizedlinearmodelswithrandomeects:AGibbssamplingapproach,"JournaloftheAmericanStatisticalAssociation,86,79{86.20