系统设计外文文献及译文 张所炜.docx

系统设计外文文献及译文 张所炜.docx

《系统设计外文文献及译文 张所炜.docx》由会员分享，可在线阅读，更多相关《系统设计外文文献及译文 张所炜.docx（19页珍藏版）》请在冰点文库上搜索。

系统设计外文文献及译文张所炜

南京工程学院

英文文献及译文

作者：

张所炜学号：

209100738

系部：

经济与管理学院

专业：

信息管理与信息系统

题目：

“投入产出分析系统”的设计与实现

指导者：

黄传峰副教授

2014年2月

Emergingchallenginginregionalinput-outputanalysis

GeoffreyJ.D.Hewings

UniversityofIllinoisatUrbana-ChampaignUSA

and

RodneyC.Jensen

UniversityofQueenslandSt.LuciAustralia

Abstract.Thechanginginterestsandfocusofresearchinthefieldofregionalinput-outputanalysisisexamined.Afterreviewingsomeoftherecenttrendsandsuggestingthetenoroftheprevailingphilosophyinthefield,attentionisfocusedonthree,interdependentemergingtrends.Thesearecharacterizedas

（1）theconceptualizationofinput-outputwithinthetraditionsofeconometricanalysis;

（2）theintegrationofinput-outputwithotherregionalandinterregionalmodelsand（3）attemptstolinkinput-outputanalysiswithregionalgrowthanddevelopmenttheory.

I.Preface

Toourknowledge,MichaelMischaikow'sresearchinterestshavenotdirectlyencompassedregionalinput-outputanalysis.However,asEditoroftheAnnals,andasahighlyrespectedstatesmaninregionalscience,hehashadasignificantinfluenceinfosteringthegrowthanddevelopmentofmanyanalyticaltechniquesinregionalanalysisincludinginput-outputanalysis.SeveralveryimportantandinfluentialarticleshaveappearedintheAnnals,manyastheresultofMischaikow'sinitiativeandencouragement.Hehasbeenafirm,committedchampionofoursub-fieldofregionalinput-outputanalysis.WearepleasedtohavetheopportunitytoofferthispaperaspartofhisFestschrift,bothasamarkofpersonalappreciationandencouragement,andtohonortheoutstandingcontributionofavaluedcolleague.

II.Introduction

Thefieldofregionalanalyticalmodellingisundergoingasignificantnewsurgeofinterestanddevelopment.Inthispaper,someofthesedevelopmentswillbereportedinthecontextofasetofemergingehallengesinthefieldofregionalinput-outputanalysis.First,however,thecurrentstateoftheartwillbereviewedbriefly.Thereafter,somegeneralcommentswillbemadeabouttheprevailingphilosophyininput-outputanalysis.Thefifthsectionofthepaperwilladdresstheemergingtrendsasawayofestablishingapossibleagendaforthefuture.Thefinalpartofthepaperprovidessomeconcludingcomments.

III.TheStateoftheArtinRegionalInput-OutputAnalysis

Withseveralrecentcontributionsunderthisumbrella（MillerandBlair,1985;Richardson,1985;HewingsandJensen,1986）,theneedforyetanothercomprehensivereviewofinput-outputattheregionallevelisnotahighpriority.Theobjectiveinthissectionisnottoprovidethedetailedcoveragethatthesepapersandmonographshavecontributed,butratherasummaryandoverviewofgeneraltrendsanddirections.Thisoverviewisintendedtofacilitatethediscussioninlatersectionsofthepaper.

Twoimportantpointsneedtobeestablishedasaprefacetoanoverviewofimportantdevelopmentsininput-output.First,whileregionalinput-outputmodelshavebecomeanacceptedandmuch-usedpartofthearsenalofanalyticaltechniques,thereisastrongsuspicionthatmanyanalystshaveahigherlevelofawarenessofthemodel'slimitationsthantheyhaveofitsutility.Thiscommentismadeonthebasisofcasualempiricismderivedfromreferees'commentsoninput-outputpaperssubmittedtojournalsandcommentarymadeonsimilarpaperspresentedatprofessionalmeetings.Weseethisprimarilyasaresultofthecontinuedhighlevelofdebate,evaluation,andtestingwhichhascharacterizedthefieldofregionalinput-output,probablyfarmorethanmostotherfieldsofeconomicanalysis.Thishealthydebateandcriticalintrospectionhasbeeninthetraditionsofanacademicandprofessionalenvironmentaimedatcontinuedimprovementandevaluationofexistinganalyticalskills.Itisnoexaggerationtosuggestthatthemostinformedandconsistentcriticsofaspectsoftheinput-outputmethodologyarethoseactivelyanddiligentlyinvolvedinresearchonthetechnique.Asissooftentheease,thenegativeaspectsofsuchactivitiestendtofUterthroughtothosewithmarginalinterestandknowledgeofthetechnique,withmoreefficiencythanpositiveaspects,creatinginthiseaseanimageofinput-outputwhichislessencouragingthanthatwarrantedbytherealityofprogress.Inasense,thissituationprovidesevidenceofaGresham'sLawofInformation--thebaddrivingoutthegood.

Second,thereisalsoanunderlyingperceptionthatinput-outputmodelshavenotadaptedwelltotheneedsofthemodernanalyst;thereappearstobesomefeelingthat"moremodern"methodsareincreasinglyrequiredinroutineanalyticalsituations,asthe"lifecycle"ofinput-outputanalysisproceedspastsomepeakofactivity.Weseesuchanattitudeasextraordinaryinlightofthecurrentunprecedentedexpansionintheuseofinput-outputattheregionallevel~wherethetechniqueisrapidlybecomingroutineinplanningandimpactstudies,andobviouslyfillinganeedtoanincreasingextent.Whileitwouldbeunfortunateifthispaperdevolvedmerelyintoadefenseofthetechnique,someoftheseperceptionsarewidelyheldtobesound.Hence,somesummarystatementofthecurrentstatusofthefieldanditsdevelopmentsappearstobeappropriateandisprovidedbelow.

Anexaminationoftheliteraturesuggeststhatanumberofimportantdevelopmentscanbehighlighted:

（1）herapidgrowthintheadoptionofinput-outputanalysis,forplanning,forecastingandgeneralimpactanalysis,attheregionallevelincountriesofallpoliticalpersuasionsandlevelsofdevelopment,tothestageofroutineapplication.Recentexperiencewouldappeartobecountertoanysuggestionthatinput-outputanalysishasreachedandpasseditszenith.Infact,inputoutputappearstobeenteringanewstageofexpandedroutineapplication.Inmanyways,thechallengesfacingtheconstructionandultimateuseofregionalinput-outputmodelsareasgreatastheyweretwoorthreedecadesago,yettheyareofadifferentkind.Ofparticularimportancehasbeenthegradualuseofthesemodelsindevelopingeconomies,particularlyinthecontextofintegratedmodels（seebelow）.

（2）thedeclineintheattentiongiventotheproductionofregionalinputoutputtables{newinput-outputtablesareappropriatelyregardedasroutineratherthansignificantevents.Nospecialfanfareisaccordedtheproductionofregionalinput-outputtablesinthedevelopedworld--unlesssomenovelaccountingscheme,datacollectionmethodorparticularapplicationhasbeenassociatedwiththeirdevelopment.Theprocesshasjbecorneroutine,acceleratedbytheavailabilityofseveralcompetingpersonalcomputerpackageswhichwillenabletheconstructionofaregionalinput-outputtablewithminimalregionaldata（forapartialreview,seeSivitanidouandPolenske,1987）.Furthermore,therehasbeenagreaterrecognitionintheliteratureofthelinkagesbetweenanumberofimportantmodellingparadigms.Thus,thebi-proportionalorRAStechniquewhichwasfirstdevelopedforupdatinginput-outputtableshasbeenshowntobepartofabroadfamilyofmatrixestimationtechniquesontheonehand（seeBatten,1982;BoyceandBatten,19861Nagurney,19871Willikens,1981）andaspecialcaseofgeneralerroranalysisinmatrixsystemsontheotherhand（seeSonisandHewings,1987）.Thisrecognitionhasleadtoagreatdealofsharedexpertiseandanenrichingoftheanalyticaltools{theflexibilityaffordedbyentropy,contingencytableandnetworkapproaches（suchasthoseofKadasandKlafsky,1976andthevariationalinequalityproposalsbyNagurney,1987）tomatrixestimationhasprovidedtheanalystwithachoiceofapproacheswhich,forthemostpart,donotdependentirelyonthedatasetavailable.

（3）amovementtowardsthedevelopmentofhybridinput-outputtables,acompromisepositionbetweengroupswhoadvocatedtheconstructionoftablesfromsurveydataaloneandthosewhosepositionisthatnonsurveytechniqueswillproducetablesoftherequisitequality.Alternativeapproachestotheconstructionofregionaltablesdominatedtheliteratureformuchofthe1970s.Whileextremepositionsweretakenintheearlieryears,therewouldappeartohavebeensignificantmellowingofopinionsandmovementstowardsthecenter,thecenterbeingdefinedastherecognitionthatpartialsurveyorhybridtableswouldbecomethedominantconstructiontechnique.Inpart,thiscompromisewasreachedonbothpragmaticandanalyticgrounds.Intheformercase,therecognitionthatthedaysofmassiveappropriationoffundsbystate,provincialandlocalgovernmentagenciesforthedenovoconstructionofinput-outputtableswasoverpropelledresearcherstoponderthealternatives.Ontheotherhand,therewasincreasingevidencethatthecensusmentality,namelythatallentriesinaninput-outputtablehadtobeobtainedfromsurveydata,wasprobablymisplaced.Notionsofanalyticalimportancebegantoprovidethewayforacompromisewhichwouldallowtheinvestigatortomaximizethequalityoftheeffortinvolvedinanydatacollectionbyfocusingongartsofthesystemwhosedirectestimationwasdeemedcritical.Whiledebatenowcentersontheidentificationoftheseelements,fewarticleshaveappearedintherecentpastwhichhaveventuredfarfromanotionofsupportforthedevelopmentofhybrid.

（4）arecognitionthatintegratedandmorespecializedmodels,withtheinput-outputmodelasonecomponent,willbemoreimportantattheregionallevelinthecomingyears,andthedevelopmentofoperationalmodelsofthistype.WhileIsard's"ChannelsofSynthesis"chapterinMethodsofRegionalAnalysiswasconsideredby