简介:AnSMIBmodelinthepowersystems,especiallythatconceringtheeffectsofhardlimitsonbifurcations,chaosandstabilityisstudied.Parameterconditionsforbifurcationsandchaosintheabsenceofhardlimitsarecomparedwiththoseinthepresenceofhardlimits.Ithasbeenprovedthathardlimitscanaffectsystemstability.Wefindthat(1)hardlimitscanchangeunstableequilibriumintostableone;(2)hardlimitscanchangestabilityoflimitcyclesinducedbyHopfbifurcation;(3)persistenceofhardlimitscanstabilizedivergenttrajectorytoastableequilibriumorlimitcycle;(4)HopfbifurcationoccursbeforeSNbifurcation,sothesystemcollapsecanbecontrolledbeforeHopfbifurcationoccurs.Wealsofindthatsuitablelimitingvaluesofhardlimitscanenlargethefeasibilityregion.Theseresultsarebasedontheoreticalanalysisandnumericalsimulations,suchasconditionforSNBandHopfbifurcation,bifurcationdiagram,trajectories,Lyapunovexponent,Floquetmultipliers,dimensionofattractorandsoon.
简介:ItiswellknownthatsplinesmoothingestimatorrelatestotheBayesianestimateunderpartiallyinformativenormalprior.Inthispaper,wederivetheconditionsfortheproprietyoftheposteriorinthenonparametricmixedefectsmodelunderthisclassofpartiallyinformativenormalpriorforfxedefectwithinversegammapriorsonthevariancecomponentsandhierarchicalpriorsforcovariancematrixofrandomefect,thenweexploretheGibbssamplingprocedure.
简介:Inthispaper,theconceptofdistributioneffectisproposedwithoutthecausaldiagram.FollowingthenotationofStone[11],weassumethattheexposuretreatmentXisanunknowndeterministicfunctionoftheconfoundersetPa(X)andarandomerrorε.Wediscusssufficientandnecessaryconditionsforhomogeneity,collapsibilityandnonconfoundingfordistributioneffectsanddiscussrelationsamongthem.
简介:ThispaperpresentsaunifieddiagnosticmethodforexponentialnonlinearmodelswithrandomeffectsbaseduponthejointlikelihoodgivenbyRobinsonin1991.Theauthorsshowthatthecasedeletionmodelisequivalenttomeanshiftoutliermodel.Fromthispointofview,severaldiagnosticmeasures,suchasCookdistance,scorestatisticsarederived.ThelocalinfluencemeasureofCookisalsopresented.Numericalexampleillustratesthatourmethodisavailable.
简介:Ageometricframeworkisproposedfornonlinearmodelswithrantiomeffects,basedonamodifiedversionofthegeometricstructurepresentedbyBates&Wattes(1980).Thisgeometricframeworkisusedtostudysomeasymptoticinferenceintermsofcurvaturesfornonlinearregressionmodelswithrandomeffects.Severalpreviousresultsareextended.
简介:Westudyamodelforthelong-termbehaviorofasingle-speciespopulationwithsomedegreeofpollutiontoleranceinapollutedenvironment.Themodelconsistsofthreeordinarydifferentialequations:oneforthepopulationdensity,onefortheamountoftoxicantinsidethelivingorganisms,andonefortheamountoftoxicantintheenvironment.Wederivesufficientconditionsforthepersistenceandtheextinctionofthepopulationdependingontheexogenousinputrateofthetoxicantintotheenvironmentandthelevelofpollutiontoleranceoftheorganisms.Numericalsimulationsarecarriedouttoillustrateourmainresults.
简介:Quantileregressionisgraduallyemergingasapowerfultoolforestimatingmodelsofconditionalquantilefunctions,andthereforeresearchinthisareahasvastlyincreasedinthepasttwodecades.Thispaper,withthequantileregressiontechnique,isthefirstcomprehensivelongitudinalstudyonmathematicsparticipationdatacollectedinAlberta,Canada.Themajoradvantageoflongitudinalstudyisitscapabilitytoseparatetheso-calledcohortandageeffectsinthecontextofpopulationstudies.Oneaimofthispaperistostudywhetherthefamilybackgroundfactorsalterperformanceonthemathematicalachievementofthestrongeststudentsinthesamewayasthatofweakerstudentsbasedonthelargelongitudinalsampleof2000,2001and2002mathematicsparticipationlongitudinaldataset.Theinterestingfindingssuggestthattheremaybedifferentialfamilybackgroundfactoreffectsatdifferentpointsinthemathematicalachievementconditionaldistribution.