1、本科毕业论文外文翻译出处SOUTHERNAGRICULTURALECONOMICSASSOCIATION20091120作者CHALINONDO,MULUGETASKAHSAI,PETERVSCHAEFFER原文ENERGYCONSUMPTIONANDECONOMICGROWTHEVIDENCEFROMCOMESACOUNTRIESABSTRACTTHISSTUDYAPPLIESPANELDATATECHNIQUESTOINVESTIGATETHELONGRUNRELATIONSHIPBETWEENENERGYCONSUMPTIONANDGDPFORAPANELOF19AFRICANCOUNT
2、RIESCOMESABASEDONANNUALDATAFORTHEPERIOD19802005INTHEFIRSTSTEP,WEEXAMINETHEDEGREEOFINTEGRATIONBETWEENGDPANDENERGYCONSUMPTIONANDFINDTHATTHEVARIABLESAREINTEGRATEDOFORDERONEINTHESECONDSTEP,WEINVESTIGATETHELONGRUNRELATIONSHIPBETWEENENERGYCONSUMPTIONANDGDPOURRESULTSPROVIDESTRONGEVIDENCETHATGDPANDENERGYCON
3、SUMPTIONMOVETOGETHERINTHELONGRUNINTHETHIRDSTEP,WEESTIMATETHELONGRUNRELATIONSHIPANDTESTFORCAUSALITYUSINGPANELBASEDERRORCORRECTIONMODELSANDFINDALONGRUNBIDIRECTIONALRELATIONSHIPBETWEENGDPANDENERGYCONSUMPTIONFURTHER,OURANALYSESREVEALTHATCAUSATIONRUNSFROMENERGYCONSUMPTIONTOGDPFORLOWINCOMECOMESACOUNTRIES1
4、INTRODUCTIONDESPITEBEINGENDOWEDWITHANARRAYOFNATURALENERGYRESOURCES,SUCHASCOAL,WATER,OIL,NATURALGAS,ANDURANIUM,SUBSAHARANAFRICASSAHASTHELOWESTPERCAPITAENERGYCONSUMPTIONLEVELSINTHEWORLDUNITEDNATIONSECONOMICCOMMISSIONOFAFRICA,2004MORETHAN80PERCENTOFTHESSAPOPULATIONRELIESONTRADITIONALENERGYSOURCES,SUCHA
5、SBIOMASS,AGRICULTURALRESIDUES,ANDOTHERPRIMITIVEENERGYSOURCES,WHICHEXACERBATEENVIRONMENTALDEGRADATIONANDAIRPOLLUTIONRELATEDHEALTHIMPACTSLEGROSETAL2009THEINADEQUATEPROVISIONOFMODERNENERGYSERVICESINSSAHASBEENCITEDBYTHEUNITEDNATIONSECONOMICCOMMISSIONFORAFRICAUNECA,2004ASALIMITINGFACTORINECONOMICGROWTHAN
6、DPOVERTYALLEVIATIONEFFORTSFOLLOWINGTHEINDEPENDENCEOFMOSTAFRICANCOUNTRIESBYTHEEARLY1970S,AFRICANLEADERSEMBRACEDREGIONALINTEGRATIONASACENTRALELEMENTOFTHEIRDEVELOPMENTSTRATEGIESWORLDENERGYCOUNCIL,2005THEPERIODMARKEDTHEBEGINNINGOFTHEFORMATIONOFREGIONALECONOMICCOMMUNITIESRECSINAFRICATHEREGIONALECONOMICCO
7、MMUNITIESWEREPRIMARILYAIMEDATPROMOTINGUNITY,ENHANCINGSUSTAINABLEDEVELOPMENT,INCREASINGCOMPETITIVENESS,ANDINTEGRATINGAFRICANCOUNTRIESINTOTHEGLOBALECONOMYTHROUGHMUTUALCOOPERATIONAMONGMEMBERCOUNTRIESOURSTUDYREGION,THECOMMONMARKETFOREASTERNANDSOUTHERNAFRICACOMESA,WHICHISCOMPOSEDOF19COUNTRIES,WASFORMEDWI
8、THTHEOBJECTIVEOFPROMOTINGREGIONALINTEGRATIONTHROUGHTRADEDEVELOPMENTWITHINCOMESA,THEREAREMARKEDDIFFERENCESINTHELEVELSOFDEVELOPMENT,NATURALENERGYRESOURCEENDOWMENT,ANDENERGYDEMANDCOGNIZANTOFTHECOMPETITIVEADVANTAGESTHATSOMEMEMBERSTATESHAVE,COMESAHASDEVELOPEDPROTOCOLSTHATPROVIDEFORCOOPERATIONINENERGYDEVE
9、LOPMENTTHROUGHTHEPOOLINGOFENERGYRESOURCESINPRINCIPLE,THESEPROTOCOLSAREAIMEDATINCREASINGENERGYACCESSIBILITYANDPROMOTINGECONOMICGROWTHTHEDIRECTIONOFCAUSATIONBETWEENENERGYCONSUMPTIONANDECONOMICGROWTHHASIMPORTANTPOLICYIMPLICATIONSFORCOMESACOUNTRIES,WHICHPURSUETHECOMMONGOALOFINCREASINGENERGYSUPPLYTHROUGH
10、REGIONALENERGYCOOPERATIONANDTRADEUNDERSTANDINGTHECAUSALRELATIONSHIPBETWEENENERGYCONSUMPTIONANDECONOMICGROWTHWILLHELPPOLICYMAKERSFORMULATEENERGYPOLICIESFORCOMESAANDITSMEMBERCOUNTRIESGIVENTHATNOATTEMPTHASBEENMADEINTHEEMPIRICALLITERATURETOQUANTIFYTHECAUSALRELATIONSHIPBETWEENENERGYCONSUMPTIONANDECONOMIC
11、GROWTHFORANYREGIONALECONOMICCOMMUNITYINAFRICA,THISSTUDYAIMSTOFILLTHATGAPBYEMPLOYINGPANELUNITROOTTESTS,PANELCOINTEGRATIONTESTS,ANDTHEDYNAMICPANELERRORCORRECTIONMODELONAPANELOFTHE19COMESACOUNTRIESTODATE,THEFEWCAUSALITYSTUDIESTHATHAVEBEENCONDUCTEDAREBASEDONINDIVIDUALCOUNTRIESANDUSETIMESERIESDATAAKINLO,
12、2008JUMBE,2004ODHIAMBO,2009WOLDERUFAEL,2006THERESTOFTHEPAPERISORGANIZEDASFOLLOWSSECTION2PROVIDESASUMMARYOFTHEECONOMICANDENERGYPROFILEOFCOMESACOUNTRIESSECTION3PRESENTSTHELITERATUREREVIEW,SECTION5PROVIDESADISCUSSIONOFTHEEMPIRICALRESULTS,ANDSECTION6CONTAINSCONCLUSIONSANDPOLICYRECOMMENDATIONS2LITERATURE
13、REVIEWINTERESTINTHECAUSALRELATIONSHIPBETWEENENERGYCONSUMPTIONANDECONOMICGROWTHWASSPAWNEDBYKRAFTANDKRAFTS1978SEMINALWORKEMPIRICALAPPROACHESTOTESTTHECAUSALRELATIONSHIPSBETWEENENERGYCONSUMPTIONANDECONOMICGROWTHHAVEBEENSYNTHESIZEDINTOFOURTESTABLEHYPOTHESESAPERGISANDPAYNE,2009THEFIRSTHYPOTHESISISTHATENER
14、GYCONSUMPTIONISAPREREQUISITEFORECONOMICGROWTHGIVENTHATENERGYISADIRECTINPUTINTHEPRODUCTIONPROCESSANDANINDIRECTINPUTTHATCOMPLEMENTSLABORANDCAPITALINPUTSEBOHON,1996TOMANANDJEMELKOVA,2003INTHISCASEAUNIDIRECTIONALGRANGERCAUSALITYRUNNINGFROMENERGYCONSUMPTIONTOGDPMEANSTHATTHECOUNTRYSECONOMYISENERGYDEPENDEN
15、T,ANDTHATPOLICIESPROMOTINGENERGYCONSUMPTIONSHOULDBEADOPTEDINTOSTIMULATEECONOMICGROWTHBECAUSEINADEQUATEPROVISIONOFENERGYMAYLIMITECONOMICGROWTHTHESECONDHYPOTHESISASSERTSTHATWHENCAUSALITYRUNSFROMECONOMICGROWTHTOENERGYCONSUMPTION,ANECONOMYISLESSENERGYDEPENDENT,ANDTHUSENERGYCONSERVATIONPOLICIES,SUCHASPHA
16、SINGOUTENERGYSUBSIDIESMAYNOTADVERSELYAFFECTECONOMICGROWTHMEHRA,2006FERGUSONETAL2000FINDSTRONGEVIDENCETHATANINCREASEINWEALTHISPOSITIVELYRELATEDTOENERGYCONSUMPTIONROSENBERG1998PROVIDESANECDOTALEVIDENCETHATINCREASEDENERGYPROVISIONPLAYEDANIMPORTANTROLEINTHEDEVELOPMENTPROCESSOFINDUSTRIALIZEDCOUNTRIESTHET
17、HIRDHYPOTHESISASSUMESTHATTHEREISNOCAUSALITYBETWEENENERGYCONSUMPTIONANDECONOMICGROWTHALSOKNOWNASTHENEUTRALHYPOTHESISTHUS,POLICIESAIMEDATCONSERVINGENERGYWILLNOTRETARDECONOMICGROWTHASAFUADAYE,2000JUMBE,2004FINALLY,THEFOURTHHYPOTHESISASSUMESABIDIRECTIONALRELATIONSHIPBETWEENENERGYCONSUMPTIONANDECONOMICGR
18、OWTHTHEIMPLICATIONOFTHEBIDIRECTIONALRELATIONSHIPISTHATENERGYCONSUMPTIONANDECONOMICGROWTHARECOMPLEMENTARY,ANDTHATANINCREASEINENERGYCONSUMPTIONSTIMULATESECONOMICGROWTH,ANDVICEVERSACHENETAL2007ALSOEMPLOYADYNAMICPANELERRORCORRECTIONMODELONAPANELOF10ASIANDEVELOPINGCOUNTRIESRESULTSFROMCHENETALINDICATEABID
19、IRECTIONALRELATIONSHIPBETWEENELECTRICITYCONSUMPTIONANDECONOMICGROWTHINTHELONGRUN,WHILECAUSALITYRUNSFROMELECTRICITYCONSUMPTIONTOECONOMICGROWTHONLYINTHESHORTRUNAPERGISANDPAYNE2009,2010EXAMINETHECAUSALRELATIONSHIPBETWEENENERGYCONSUMPTIONANDECONOMICGROWTHFORAPANELOF11COUNTRIESOFTHECOMMONWEALTHOFINDEPEND
20、ENTSTATESIITHEYFINDUNIDIRECTIONALCAUSATIONFROMENERGYCONSUMPTIONTOECONOMICGROWTHINTHESHORTRUN,ANDABIDIRECTIONALRELATIONSHIPBETWEENENERGYCONSUMPTIONANDGROWTHOFREALOUTPUTINTHELONGRUNINGENERALTHEEMPIRICALLITERATURESHOWSTHATENERGYCONSUMPTIONSTIMULATESECONOMICGROWTH,ANDVICEVERSA3METHODOLOGYANDDATAPREVIOUS
21、STUDIESHAVEEXAMINEDTHERELATIONSHIPBETWEENENERGYCONSUMPTIONELECTRICITYCONSUMPTIONANDECONOMICGROWTHINSUBSAHARANAFRICAUSINGCOUNTRYLEVELDATAANDTIMESERIESTECHNIQUESINTHISSTUDY,WEEMPLOYPANELESTIMATIONTECHNIQUESTODETERMINETHEDYNAMICRELATIONSHIPBETWEENENERGYCONSUMPTIONANDECONOMICGROWTHTHEMETHODOLOGYADOPTEDI
22、NTHISSTUDYUSESATHREESTEPPROCEDUREFIRST,PANELUNITROOTTESTSAREAPPLIEDTOTESTTHEDEGREEOFINTEGRATIONBETWEENECONOMICGROWTHANDENERGYCONSUMPTIONSECOND,PANELCOINTEGRATIONTECHNIQUESPEDRONI,1999AREAPPLIEDTODETERMINETHELONGRUNRELATIONSHIPBETWEENENERGYCONSUMPTIONANDGDPFINALLY,ADYNAMICPANELERRORCORRECTIONMODELISA
23、PPLIEDTODETERMINETHEDIRECTIONOFCAUSATIONINTHESHORTRUNANDLONGRUNPANELUNITROOTTESTSPANELUNITROOTTESTSAREUSEDTOEXAMINETHEDEGREEOFINTEGRATIONBETWEENGDPANDENERGYCONSUMPTIONSUCHTESTSHAVEBEENSUGGESTEDASANALTERNATIVEFOREXAMININGTHECAUSALRELATIONSHIPBETWEENENERGYCONSUMPTIONANDECONOMICGROWTHINAPANELFRAMEWORKT
24、HISESTIMATIONMETHODISBECOMINGMOREPOPULARBECAUSETHEASYMPTOTICDISTRIBUTIONISSTANDARDNORMAL,INSTEADOFNONNORMALASYMPTOTICDISTRIBUTIONSBALTAGI,2004WETESTFORUNITROOTSUSINGTHREEPANELBASEDMETHODSPROPOSEDBYLEVIN,LINANDCHU2002,HEREAFTERREFERREDTOASLLC,IM,PESARAN,ANDSHIN2003,HEREAFTERREFERREDTOASIPS,ANDHADRI20
25、00FOREACHESTIMATIONTECHNIQUE,WETESTFORUNITROOTSINTHEPANELUSINGTWOTYPESOFMODELSIIITHEFIRSTMODELINVOLVESESTIMATINGTHEVARIABLESINLEVELFORMWITHANDWITHOUTADETERMINISTICTREND,WHILETHESECONDMODELINVOLVESESTIMATINGTHEFIRSTDIFFERENCEOFTHEVARIABLESWITHANDWITHOUTADETERMINISTICTRENDTHELLCTESTISTHEMOSTWIDELYUSED
26、PANELUNITROOTTESTANDCANBESPECIFIEDASFOLLOWSIPJITJITIITIIITEYPYY111ISTHEFIRSTDIFFERENCEOPERATOR,ITYISTHESERIESOFOBSERVATIONSFORCOUNTRYIFORT1,TTIMEPERIODSTHETESTHASTHENULLHYPOTHESIS00IHFORALLIAGAINSTTHEALTERNATIVEOF10IHFORALLI,WHICHPRESUMESTHATALLSERIESARESTATIONARYLLCASSUMESTHATISHOMOGENOUSACROSSREGI
27、ONSANDTHETESTISBASEDONTHETBARSTATISTICTHEIPSTESTISANEXTENSIONOFTHELLCTESTANDISBASEDONTHEMEANOFTHEINDIVIDUALUNITROOTSTATISTICINTHESAMEMODELUSEDINTHELLCTESTUNLIKETHELLCTEST,THEIPSTESTALLOWSFORHETEROGENEITYINTHEVALUEOFUNDERTHEALTERNATIVEHYPOTHESISTHEHADRITESTISANLMBASEDTESTWITHTHENULLHYPOTHESISTHATALLS
28、ERIESINTHEPANELARESTATIONARYPANELCOINTEGRATIONTHESECONDSTEPOFOUREMPIRICALWORKINVOLVESINVESTIGATINGTHELONGRUNRELATIONSHIPBETWEENENERGYCONSUMPTIONANDGDP,USINGTHEPANELCOINTEGRATIONTECHNIQUEDUETOPEDRONI1999THISTECHNIQUEALLOWSFORHETEROGENEITYAMONGINDIVIDUALMEMBERSOFTHEPANELANDISANIMPROVEMENTOVERCONVENTIO
29、NALCOINTEGRATIONTESTSFOLLOWINGPEDRONISMETHODOLOGY,THECOINTEGRATIONRELATIONSHIPWEESTIMATEISSPECIFIEDASFOLLOWSITITITIITLECLGDP2LECANDLGDPARETHENATURALLOGARITHMSOFTHEOBSERVABLEVARIABLES,T1,TARETIMEPERIODSI1,NAREPANELMEMBERSIDENOTESCOUNTRYSPECIFICEFFECTS,TISTHEDETERMINISTICTIMETRENDS,ANDITISTHEESTIMATED
30、RESIDUALTHEESTIMATEDRESIDUALINDICATESTHEDEVIATIONFROMTHELONGRUNRELATIONSHIPWITHTHENULLOFNOCOINTEGRATION,THEPANELCOINTEGRATIONISESSENTIALLYATESTOFUNITROOTSINTHEESTIMATEDRESIDUALSOFTHEPANELPEDRONI1999SHOWSTHATTHEREARESEVENDIFFERENTSTATISTICSFORTHECOINTEGRATIONTESTTHEYARETHEPANELVSTATISTIC,PANELSTATIST
31、IC,PEDRONIPANELPPSTATISTIC,PANELAUGMENTEDDICKEYFULLERADFSTATISTIC,GROUPRHOSTATISTIC,GROUPPPSTATISTIC,ANDGROUPADFSTATISTICTHEFIRSTFOURSTATISTICSAREKNOWNASPANELCOINTEGRATIONSTATISTICSANDAREBASEDONTHEWITHINAPPROACHTHELASTTHREESTATISTICSAREGROUPPANELCOINTEGRATIONSTATISTICSANDAREBASEDONTHEBETWEENAPPROACH
32、INTHEPRESENCEOFACOINTEGRATINGRELATIONSHIP,THERESIDUALSAREEXPECTEDTOBESTATIONARYTHEPANELVTESTISAONESIDEDTEST,WITHTHENULLOFNOCOINTEGRATIONBEINGREJECTEDWHENTHETESTHASALARGEPOSITIVEVALUETHEOTHERTESTSREJECTTHENULLHYPOTHESISOFNOCOINTEGRATIONWHENTHEYHAVELARGENEGATIVESTATISTICSDATADATAUSEDINTHISANALYSISAREP
33、OOLEDANNUALTIMESERIESFORNOMINALGDPHEREAFTERREFERREDTOASGDPANDENERGYCONSUMPTIONECHEREAFTERFOR19COMESACOUNTRIESFORTHEPERIOD1980TO2005BTUOFENERGYISUSEDASAPROXYFORENERGYCONSUMPTIONEC,ANDTHISDATAISOBTAINEDFROMUNITEDSTATESENERGYINFORMATIONADMINISTRATIONEIAGDPDATACOMEFROMTHEINTERNATIONALMONETARYFUNDIMFWORL
34、DECONOMICOUTLOOK2008ALLVARIABLESUSEDINTHEESTIMATIONAREINNATURALLOGARITHMFORM4CONCLUSIONSANDPOLICYRECOMMENDATIONSTHEPURPOSEOFTHISSTUDYWASTOTESTFORGRANGERCAUSALITYBETWEENENERGYCONSUMPTIONANDGDPINCOMESACOUNTRIESUSINGPANELCAUSALITYTESTSFROMTHETESTRESULTS,WECONCLUDETHATINTHESHORTRUNTHENEUTRALHYPOTHESISHO
35、LDS,BUTINTHELONGRUN,THEREISSTRONGCAUSATIONRUNNINGINBOTHDIRECTIONSFORTHE19COUNTRIESINOURSTUDYINLOWINCOMECOMESACOUNTRIES,THEREISASHORTRUNCAUSATIONTHATRUNSFROMENERGYCONSUMPTIONTOGDPFROMTHEFOREGOING,ITCANBEINFERREDTHATPOLICIESTHATSTIMULATEBOTHENERGYCONSUMPTIONANDGDPGROWTHSHOULDBEFORMULATEDANDIMPLEMENTED
36、ITISREASONABLETOCONCLUDETHATONEFACTOREXPLAININGCOMESACOUNTRIESPOORECONOMICGROWTHISTHELACKOFINVESTMENTSINENERGYINFRASTRUCTUREANDSERVICESTHUS,THECURRENTLOWINVESTMENTINENERGYINFRASTRUCTUREMAYBEANOBSTACLETHATMAYPREVENTSOMECOMESAMEMBERSTATESFROMREACHINGTHEMILLENNIUMDEVELOPMENTGOALSASACONSEQUENCE,ENERGYRE
37、LATEDPROBLEMSAREANDWILLBECRUCIALPOLICYISSUESFORCOMESACOUNTRIESAGAINSTTHISBACKGROUND,RELYINGONVOLATILEENERGYMARKETSWILLNOTGUARANTEESUSTAINABLEDEVELOPMENTANDGREATERREGIONALENERGYSELFSUFFICIENCYSHOULDBEONEOFTHEMAJOROBJECTIVESOFCOMESACOUNTRIESTHESIGNIFICANTHYDROELECTRICANDGEOTHERMALPOTENTIALS,ANDTHEPROV
38、ENOILANDGASRESERVESINCOMESACOUNTRIESCANBETAPPEDTORELIABLYSUPPLYLOWCOSTENERGYTOTHEREGIONANDTHENIMPROVEENERGYSUPPLY,INGENERAL译文能源消耗,经济增长以东南非共同市场国家为例摘要本文运用面板数据技术,基于19个非洲国家(东南非共同市场)19802005年期间每年的数据,来探讨能源消费与国内生产总值的长期运行关系。首先,研究国内生产总值和能源消耗一体化的程度,发现变量有秩序的整合在一起。然后,探讨能源消费与GDP的长期关系,研究提供了强有力的证据表明国内生产总值和能源消耗在长期运
39、行中将同时发生变化。最后,使用面板的错误校正模型来预测能源消费和GDP的长期运行关系以及他们的因果关系测试,并找到国内生产总值和能源消费的长期双向关系。此外,分析表明,能源消费与GDP的因果关系适用于东南非共同市场的低收入国家。一、引言撒哈拉以南的非洲国家(SSA)是世界上人均能源消费水平最低的国家(联合国非洲经济委员会,2004),尽管被赋予了如煤,水,石油,天然气和铀等一系列的自然能源。80以上的撒哈拉以南的非洲国家依赖传统能源,例如,生物,农业废弃物和其他原始能源,这些加剧了环境恶化和空气污染相关的健康影响(LEGROSETAL2009)。现代能源服务的不充分供给已被联合国经济委员会(U
40、NECA,2004)引用作为经济增长和扶贫工作的限制性因素。继非洲20世纪70年代以来,大多数非洲独立国家的领导人把区域一体化作为其发展战略的核心要素(世界能源理事会,2005)。这一时期标志着非洲区域经济共同体开始形成(RECS)。区域经济共同体的主要目的是促进团结,加强可持续发展,提高竞争力和促进非洲成员国通过互相合作进入全球经济一体化。我们的研究区域是由19个国家组成的东南非共同市场,它成立的目标是为了通过贸易的发展促进区域一体化。在东南非共同市场,自然资源禀赋和能源的需求都有着显著的发展水平差异。认识到一些成员国有竞争优势,东南非共同市场通过汇集能源来提供能源发展合作方面的开发协议。原
41、则上,这些协议旨在提高能源的使用水平和促进经济增长。能源消费与经济增长之间因果关系的方向对东南非共同市场的国家具有重要的政策含义,其追求的共同目标是通过区域能源合作和贸易来增加能源的供给。了解经济增长和能源消费之间的因果关系将帮助决策者制定东南非共同市场成员国的能源政策。鉴于还没有取得关于非洲任何一个区域经济共同体的实证文献来量化能源消费与经济增长因果关系,这个研究旨在纠正以这19个东南非国家为一组的采用面板单位根检验,面板协整检验,动态面板误差修正模型之间的误差。迄今,只有少数几个基于个别国家利用时间序列数据(AKINLO,2008JUMBE,2004ODHIAMBO,2009沃尔德RUFA
42、EL,2006)来进行因果关系研究。本文的其余部分组织如下,第2部分介绍了文献回顾,第3部分主要是方法论和数据来源。第4部分得出结论并提出政策建议。二、文献回顾研究经济增长和能源消费因果关系的兴趣是由KRAFT和KRAFTS(1978)激发的。用实证的方法来检验能源消耗与经济增长之间的因果关系已被综合成四种可检验的假设(APERGISANDPAYNE,2009。第一个假设是能源消耗是经济增长的先决条件,因为能源是生产过程中的一种直接投入和劳动力和资本补充的间接投入(EBOHON,1996TOMANANDJEMELKOVA,2003)。在这种情况下,一种从能源消费到GDP的单向格兰杰因果关系意味
43、着该国的经济增长依赖能源消耗,而促进能源消费的政策应该被采用来刺激经济增长,因为能源供应不足可能限制经济增长。第二种假设断言,当因果关系从经济增长到能源消费,一个经济较少依赖能源的,从而采用能源保护政策,例如逐步取消能源补贴,可能不会影响经济增长(MEHRA,,2006)。FERGUSONETAL(2000)发现强有力的证据表明,财富的增加和能源消耗呈正相关。ROSENBERG(1998)提供证据表明能源供给的增加在工业化国家的发展过程中发挥了重要作用。第三个假说假定能源消费和经济增长之间没有因果关系(也称为中性的假设)。因此,制定政策的目的是保护能源不会阻碍经济增长(ASAFUADAYE,2
44、000JUMBE,2004)。最后,第四种假设假定经济增长和能源消耗具有双向关系。蕴涵的双向关系是指能源消耗和经济增长是相辅相成的,能源消耗的增加能刺激经济的增长,反之亦然。CHENETAL(2007)采用动态面板误差纠正模型对10个亚洲发展中国家进行研究,结果表明在长期运行中电力消耗和经济增长有一种双向关系,而电力消费和经济增长的因果关系只存在于短期运行之中。APERGIS和PAYNE(2009,2010)利用独立国家联合体中的11个国家为一组研究了能源消费与经济增长的因果关系。他们发现能源消耗和经济增长的单向因果关系存在于短期,而能源消耗和实际产出增长的双向因果管子存在于长期。从一般实证文
45、献可得,能源消耗刺激经济增长,经济增长促进能源消耗。三、方法论和数据以前的研究曾利用国家级数据和时间系列技术探讨了撒哈拉沙漠以南地区能源消耗(电力消耗)和经济增长之间的关系。在这项研究中,我们采用面板估算技术来确定动态能源消耗和经济增长之间的关系。在本文中采用的方法论是三步骤程序。首先,用面板单位根检验被来检验经济增长和能源消耗之间一体化的程度。其次,用面板协整技术(PEDRONI,1999)来决定能源消耗与国内生产总值之间的长期运行关系。最后,用动态面板误差修正模型来决定长期和短期运行的因果关系的方向。面板单位根检验面板单位根检验用来检验能源消耗和国内生产总值一体化的程度。这种检验已被建议在
46、平板框架里作为一种检验能源消耗和经济增长因果关系的一种选择。这种估计方法会变得越来越流行因为它是呈标准渐近正态分布的,而不是呈非标准的渐近正态分布(BALTAGI,2004)。我们利用LEVIN,LIN和CHU(2002)提出的用三种基于面板的方法来测试单位根,以下简称为LLC,IM,PESARAN,ANDSHIN2003,以下简称IPS和HADRI(2000)。对每一种估计方法而言,我们使用两种不同类型的模型来测试单位面板。第一个模型,包含在同一形式上没有一个特定趋势的可变估计量。而第二个模型包含第一个差异变量的估计,没有特定趋势。LCC检验是使用最广泛的面板单位根检验,表达式如下ITJIT
47、PJIITIEYPYYIIIT11是一阶差分因子,ITY是国家I在T(T1,2,T)中一系列的观察值。这个检验的零假设是0HI0和IHI0,假设这个序列是固定不变的。LLC假定在特定的区域里是均匀分布的,所用的检验方法是T检验。IPS检验是LLC检验的延伸,它以使用和LLC检验一样的模型来进行个别单位根统计的方法为基础。与LLC检验不同的是,IPS检验允许异质性在值下备择假设。HADRI检验是所有序列是固定不变的零假设,基于LM检验。面板协整第二步是我们的实证研究经验主要包括调查能源消费与国内生产总值长期运系,使用的是PEDRONI(1999)的面板协整方法,这种方法适用于个别面板小组的异质性
48、的并改进了传统的协整检验。根据佩德罗尼的方法论,我们估计协整关系的公式如下ITITITIITLECLGDPLEC和LGDP是可观测变量的自然对数,T的时间取值是1,2,T,I的面板成员的取值范围是1,2,N,I表示特定国家的影响。T具有确定的时间趋势,IT是剩余估计量。剩余估计量表示在长期运行关系中的偏差,面板协整是面板剩余量一种必要的单位根检验。PEDRONI(1999)表示有7种不同的协整检验统计法,有面板V统计,面板统计量,PEDRONIPANEL(PP)统计量,PANELAUGMENTEDDICKEYFULLER(ADF)统计量,小组RHO统计量,小组PP统计量和小组ADF统计量。前四
49、种统计方法可以被认为是在给定范围内面板协整统计,最后三组统计是基于方法范围内的小组面板协整统计。在出现协整关系的时候,期望残差是平稳变化的。面板V检验是用非协整检验出现一个很大的正数的时候将会被拒绝的一种单一的检验,其他检验是在没有协整关系的零假设出现一个很大的负数时被拒绝。数据收集从19802005年东南非共同市场的19个国家每年的名义国内生产总值(以下简称GDP)和能源消耗(以下简称EC)的数据进行分析。BTU能源被用来看做是能源消耗(EC)的一种代表,这些数据来源于美国能源情报署(EIA)。GDP数据来自于国际货币基金组织(IMF)的世界经济展望2008。在估计中使用的所有的变量都是自然数的对数形式。四结论和政策建议这篇文章的研究目的是为了用面板因果检验来证明东南非共同市场国家能源和国内生产总值间的GRANGER因果关系。的使用面板因果检验。从检验的结果来看,我们可以得出结论,短期中性假设成立,但从长远来看,在我们研究19个国家中存在很强的双向因果关系。在低收入的东南非共同市场国家,能源消耗和国内生产总值在一个短期内具有因果关系。从前面可以得出,我们应该尽快制定和实施能同时刺激了能源消费与GDP增长的政策。解释东南非共同市场经济增长