K. G. Troitzsch
{"title":"统计显著性检验在仿真结果分析中的意义","authors":"K. G. Troitzsch","doi":"10.4018/IJATS.2016010102","DOIUrl":null,"url":null,"abstract":"Thisarticlediscussesthequestionofwhethersignificancetestsonsimulationresultsaremeaningful atall.Itisalsoarguedthatitistheeffectsizemuchmorethantheexistenceoftheeffectiswhat matters.Itisthedescriptionofthedistributionfunctionofthestochasticprocessincorporatedinthe simulationmodelwhichisimportant.Thisisparticularlywhenthisdistributionisfarfromnormal, whichisparticularlyoftenthecasewhenthesimulationmodelisnonlinear.Tothisend,thisarticle usesthreedifferentagent-basedmodelstodemonstratethattheeffectsofinputparametersonoutput metricscanoftenbemade“statisticallysignificant”onanydesiredlevelbyincreasingthenumber ofruns,evenfornegligibleeffectsizes.Theexamplesarealsousedtogivehintsastohowmany runsarenecessarytoestimateeffectsizesandhowtheinputparametersdetermineoutputmetrics. KeywoRdS Cumulative Periodogram, Deduction, Microspecification, Multilevel Model, Random Number Generator, Statistical Significance Level, Stochastic Process, Validity","PeriodicalId":93648,"journal":{"name":"International journal of agent technologies and systems","volume":"25 1","pages":"18-45"},"PeriodicalIF":0.0000,"publicationDate":"2016-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":"{\"title\":\"The Meaningfulness of Statistical Significance Tests in the Analysis of Simulation Results\",\"authors\":\"K. G. Troitzsch\",\"doi\":\"10.4018/IJATS.2016010102\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Thisarticlediscussesthequestionofwhethersignificancetestsonsimulationresultsaremeaningful atall.Itisalsoarguedthatitistheeffectsizemuchmorethantheexistenceoftheeffectiswhat matters.Itisthedescriptionofthedistributionfunctionofthestochasticprocessincorporatedinthe simulationmodelwhichisimportant.Thisisparticularlywhenthisdistributionisfarfromnormal, whichisparticularlyoftenthecasewhenthesimulationmodelisnonlinear.Tothisend,thisarticle usesthreedifferentagent-basedmodelstodemonstratethattheeffectsofinputparametersonoutput metricscanoftenbemade“statisticallysignificant”onanydesiredlevelbyincreasingthenumber ofruns,evenfornegligibleeffectsizes.Theexamplesarealsousedtogivehintsastohowmany runsarenecessarytoestimateeffectsizesandhowtheinputparametersdetermineoutputmetrics. KeywoRdS Cumulative Periodogram, Deduction, Microspecification, Multilevel Model, Random Number Generator, Statistical Significance Level, Stochastic Process, Validity\",\"PeriodicalId\":93648,\"journal\":{\"name\":\"International journal of agent technologies and systems\",\"volume\":\"25 1\",\"pages\":\"18-45\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2016-01-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"1\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"International journal of agent technologies and systems\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.4018/IJATS.2016010102\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"International journal of agent technologies and systems","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.4018/IJATS.2016010102","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 1
The Meaningfulness of Statistical Significance Tests in the Analysis of Simulation Results
Thisarticlediscussesthequestionofwhethersignificancetestsonsimulationresultsaremeaningful atall.Itisalsoarguedthatitistheeffectsizemuchmorethantheexistenceoftheeffectiswhat matters.Itisthedescriptionofthedistributionfunctionofthestochasticprocessincorporatedinthe simulationmodelwhichisimportant.Thisisparticularlywhenthisdistributionisfarfromnormal, whichisparticularlyoftenthecasewhenthesimulationmodelisnonlinear.Tothisend,thisarticle usesthreedifferentagent-basedmodelstodemonstratethattheeffectsofinputparametersonoutput metricscanoftenbemade“statisticallysignificant”onanydesiredlevelbyincreasingthenumber ofruns,evenfornegligibleeffectsizes.Theexamplesarealsousedtogivehintsastohowmany runsarenecessarytoestimateeffectsizesandhowtheinputparametersdetermineoutputmetrics. KeywoRdS Cumulative Periodogram, Deduction, Microspecification, Multilevel Model, Random Number Generator, Statistical Significance Level, Stochastic Process, Validity