简介:1.IntroductionandMainResultSincethedefinitiononthecompleteconvergencewasintroducedbyHsuandRobbins[1],therehavebeenmanyauthorswhodevotethemselvestothestudyonthisconvergenceofiidrandomvariables,seeGut[2,3],BaiandSu[4]andLin[5].Meanwhile,somescholarshav...
简介:Ourobjectiveisabetterunderstandingoftheroleofphysicalpropertiesofrealfluidsinthethermodynamicsofcavitationinimpurewater.Anextensiontotheclassicalhomogenousnucleationtheorysuitableformixturesispresentedinattempttoaddressthediscrepancybetweenthetheoreticalpredictionsandpracticalobservationsofcavitationratesinwateratnormaltemperatures.Theextensiontakesintoaccoutthenon-equilibrium(dissipative)effectsinvolvedinnucleiformationthroughasubstancedependentcorrectioncoefficienttobedeterminedexperimentally.Thetheoryofthermodynamicfluctuationsisappliedtoderivetheworkofformationofabubblenucleus.Thevalueofthecorrectioncoefficientisestimatedusingpreliminaryexperimentaldatafromaconvergent-divergentnozzle.Anapplicationoftheresultstothenumericalpredictionofthecavitationzonesinaradial-flowwaterpumpisshown.
简介:TheanalysisofcelllossratesinanATMMuxwithlossprioritiesisanimportantprobleminthe-studyoftrafficcontrolinATMnetworks.Inthispaper,thelossratesofthecellswithdifferentprioritiesinanATMMuxareanalyzedbyapproximatingtheactualinputprocesswithtwo-stateMMDPandfluidflowtechnique,andtheanalyticalexpressionsoftherelationbetweenthelossratesandthebuffersizeareobtained.Simulationshowsthattheapproachissullicientlyaccurateforapplications.
简介:不同变化(替换)与微妙的还重要的差别当模特儿的为估计在编码蛋白质的序列之中的率采用的同义、非同义的替换的方法,它导致进化信息的不同估计。自从在指向的数据集以内的顺序变化的数量总是是无法预言的,很少注意都没为获得可靠估计被奉献给方法的比较。到我们的知识,几乎没有在关于这些不同方法的评估的文学可得到的很少信息。在这研究,我们比较了六个广泛地使用的方法并且用模仿的序列向结果提供了评估。结果显示那合并顺序特征(例如transition/transversion,偏爱和nucleotide/codon频率偏导)进方法能产出更好的性能。我们建议结论与有关或源于Ka,K分析不应该乐意地从一个方法仅仅根据结果被拉。
简介:ExperimentswerecarriedoutonavegetablefieldwithPekingcabbage(Brassicapekinensis(Lour.)Rupr.),cabbage(Brassicachinensisvar.OleiferaMakinoandnemoto),greencabbage(BrassicachinensisL.),spinach(SpinaciaoleraceaL.)andrape(BrassicacampestrisL.)tostudytheeffectsofNformsandNratesontheirgrowthandnitrateaccumulation.Theresultsindicatedthatapplicationofammoniumchloride,ammoniumnitrate,sodiumnitrateandureasignificantlyincreasedtheyieldsandnitrateconcentrationsofPekingcabbageandspinach.Althoughnosignificantdifferencewasfoundintheyieldsafterapplicationofthe4Nforms,nitrateNincreasednitrateaccumulationinvegetablesmuchmorethanammoniumN.ThevegetableyieldswerenotincreasedcontinuouslywithNrateincrease,andoversupplyofNreducedtheplantgrowth,leadingtoayielddecline.Thistrendwasalsotruefornitrateconcentrationsinsomevegetablesandatsomesamplingtimes.However,asawhole,thenitrateconcentrationsinvegetableswerepositivelycorrelatedwithNrates.Thus,additionofNfertilizertosoilwasthemajorcauseforincreasesinnitrateconcentrationsinvegetables.Nitrateconcentrationsweremuchhigherinroots,stemsandpetiolesthaninbladesatanyNrate.
简介:Recurrenteventdataoftenarisesinbiomedicalstudies,andindividualswithinaclustermightnotbeindependent.Weproposeasemiparametricadditiveratesmodelforclusteredrecurrenteventdata,whereinthecovariatesareassumedtoaddtotheunspecifiedbaselinerate.Fortheinferenceonthemodelparameters,estimatingequationapproachesaredeveloped,andbothlargeandfinitesamplepropertiesoftheproposedestimatorsareestablished.
简介:Recurrenteventsdatawithaterminalevent(e.g.death)oftenariseinclinicalandobservationalstudies.Mostofexistingmodelsassumemultiplicativecovariateeffectsandmodeltheconditionalrecurrenteventrategivensurvival.Inthisarticle,weproposeageneraladditive-multiplicativeratesmodelforrecurrenteventdatainthepresenceofaterminalevent,wheretheterminaleventstopthefurtheroccurrenceofrecurrentevents.Basedontheestimatingequationapproachandtheinverseprobabilityweightingtechnique,weproposetwoproceduresforestimatingtheregressionparametersandthebaselinemeanfunction.Theasymptoticpropertiesoftheresultingestimatorsareestablished.Inaddition,somegraphicalandnumericalproceduresarepresentedformodelchecking.Thefinite-samplebehavioroftheproposedmethodsisexaminedthroughsimulationstudies,andanapplicationtoabladdercancerstudyisalsoillustrated.
简介:TheeffectofQiGongreflectedbyrheoencephalographyandheartratewerestudiedinthispaper.ThefunctionalrelationshipbetweenwaveamplitudeandtheheartratesreflectedbytwoprocessesofQiGong,“JingGong”forenergyprovoca-tive,and“FaGong”forenergyreleasehavebeencomparedandanalysed.28casesofwelltrained“QiGong”practitionersvolunteerinthisstudy.Therheoencepha-
简介:Theratioofnonsynonymoussubstitutionrate(Ka)tosynonymoussubstitutionrate(Ks)iswidelyusedasanindicatorofselectivepressureatsequencelevelamongdifferentspecies,anddiversemutationmodelshavebeenincorporatedintoseveralcomputingmethods.Wehavepreviouslydevelopedanewγ-MYNmethodbycapturingakeydynamicevolutiontraitofDNAnucleotidesequences,inconsiderationofvaryingmutationratesacrosssites.WenowreportafurtherimprovementofNG,LWL,MLWL,LPB,MLPB,andYNmethodsbasedonanintroductionofgammadistributiontoillustratethevariationofrawmutationrateoversites.Thenoveltycomesintwoways:(1)weincorporateanoptimalgammadistributionshapeparameteraintoγ-NG,γ-LWL,γ-MLWL,γ-LPB,γ-MLPB,andγ-YNmethods;(2)weinvestigatehowvariablesubstitutionratesaffectthemethodsthatadoptdifferentmodelsaswellastheinterplayamongfourevolutionalfeatureswithrespecttoKa/Kscomputations.Ourresultssuggestthatvariablesubstitutionratesoversitesundernegativeselectionexhibitanoppositeeffectonωestimatescomparedwiththoseunderpositiveselection.WebelievethatthesensitivityofournewmethodshasbeenimprovedthanthatoftheiroriginalmethodsunderdiverseconditionsanditisadvantageoustointroducenovelparametersforKa/Kscomputation.
简介:Stronguniformconsistencyratesaregivenforkerneltypeestimatorsoftheconditionalfunctionwith(?)-mixingsample.Especially,fornonparametricestimatorsofkerneldensity,theregressionfunctionwhenYisbounded,conditionaldf’s,L-smoothingandM-smoothing,weobtainthesamerateO((n/logn)-1/3)asinthei.i.d.sampleestablishedbyH(?)rdle,JanssenandSerfling.