固定效應(yīng)模型及估計(jì)原理說(shuō)明

1、固定效應(yīng)模型及估計(jì)原理說(shuō)明 / 29word 范文固定效應(yīng)模型的估計(jì)原理說(shuō)明在面板數(shù)據(jù)線性回歸模型中，如果對(duì)于不同的截面或不同的時(shí)間序列，只是 模型的截距 項(xiàng)是不同的，而模型的斜率系數(shù)是相同的，則稱此模型為固定效應(yīng)模 型。固定效應(yīng)模型分為三 類：1 個(gè)體固定效應(yīng)模型個(gè)體固定效應(yīng)模型是對(duì)于不同的縱剖面時(shí)間序列（個(gè)體）只有截距項(xiàng)不同的模型：K5 ikxkit uit（1 ）k2從時(shí)間和個(gè)體上看，面板數(shù)據(jù)回歸模型的解釋變量對(duì)被解釋變量的邊際影響 均是相同 的，而且除模型的解釋變量之外，影響被解釋變量的其他所有（未包括在回歸模型或不可觀測(cè) 的）確定性變量的效應(yīng)只是隨個(gè)體變化而不隨時(shí)間變化時(shí)。檢驗(yàn)：采用

2、無(wú)約束模型和有約束模型的回歸殘差平方和之比構(gòu)造F統(tǒng)計(jì)量，以檢驗(yàn)設(shè)定個(gè)體固定效應(yīng)模型的合理性。F模型的零假設(shè)：（RRSS URSS）N1 F（N 1,N 仃1）K1）URSS/（NT N K1）RRSS是有約束模型（即混合數(shù)據(jù)回歸模型）的殘差平方和，URSS是無(wú)約束模型ANCOVA估計(jì)的殘差平方和或者LSDV估計(jì)的殘差平方和實(shí)踐：、數(shù)據(jù)：已知1996-2002年中國(guó)東北、華北、華東15個(gè)省級(jí)地區(qū)的居民家庭人均消費(fèi)（cp，不變價(jià)格）和人均收入（ip，不變價(jià)格）居民，利用數(shù)據(jù)（1 ）建立面板數(shù)據(jù)（panel data）工作文件；（2）定義序列名并輸入數(shù)據(jù)；（3）估計(jì)選擇面板模型；（4）面板單位根檢驗(yàn)

3、。年人均消費(fèi)（con sume）和人均收入（income）數(shù)據(jù)以及消費(fèi)者價(jià)格指數(shù)（p）分別見(jiàn)表1，2和3表1 1996 2002年中國(guó)東北、華北、華東15個(gè)省級(jí)地區(qū)的居民家庭人均消費(fèi)（元）數(shù)據(jù)人均消費(fèi)19961997199819992CONSUMEAH3607.433693.553777.413901.814232.984517.654736.52CONSUMEBJ5729.526531.816970.837498.488493.498922.7210284.6CONSUMEFJ4248.474935.955181.455266.695638.746015.116631.68CONSUMEHB

4、3424.354003.713834.434026.34348.474479.755069.28CONSUMEHLJ3110.923213.423303.153481.743824.444192.364462.08CONSUMEJL3037.323408.033449.743661.684020.874337.224973.88CONSUMEJS4057.54533.574889.435010.915323.185532.746042.6CONSUMEJX2942.113199.613266.813482.333623.563894.514549.32CONSUMELN3493.023719.

5、913890.743989.934356.064654.425342.64CONSUMENMG2767.843032.33105.743468.993927.754195.624859.88CONSUMESD3770.994040.634143.964515.0550225252.415596.32CONSUMESH6763.126819.946866.418247.698868.199336.110464CONSUMESX3035.593228.713267.73492.983941.874123.014710.96CONSUMETJ4679.615204.155471.015851.536

6、121.046987.227191.96CONSUMEZJ5764.276170.146217.936521.547020.227952.398713.08表2 1996 2002年中國(guó)東北、華北、華東15個(gè)省級(jí)地區(qū)的居民家庭人均收入（元）數(shù)據(jù)人均收入19961997199819992INCOMEAH4512.774599.274770.475064.65293.555668.86032.4INCOMEBJ7332.017813.168471.989182.7610349.6911577.7812463.92INCOMEFJ5172.936143.646485.636859.817432.26

7、8313.089189.36INCOMEHB4442.814958.675084.645365.035661.165984.826679.68INCOMEHLJ3768.314090.724268.54595.144912.885425.876100.56INCOMEJL3805.534190.584206.644480.0148105340.466260.16INCOMEJS5185.795765.26017.856538.26800.237375.18177.64INCOMEJX3780.24071.324251.424720.585103.585506.026335.64INCOMELN

8、4207.234518.14617.244898.615357.795797.016524.52INCOMENMG3431.813944.674353.024770.535129.055535.896051INCOMESD4890.285190.795380.085808.966489.977101.087614.36INCOMESH8178.488438.898773.110931.6411718.0112883.4613249.8INCOMESX3702.693989.924098.734342.614724.115391.056234.36INCOMETJ5967.716608.3971

9、10.547649.838140.58958.79337.56INCOMEZJ6955.797358.727836.768427.959279.1610464.6711715.6固定效應(yīng)模型及估計(jì)原理說(shuō)明 / 29word 范文Workfi e CreateWorkfile structure typeDa ted 一 regular frequency virregular Dated and Panel workfiles may b 皂 made from LinstrLJCtLJred orkfile by later specifying date and/or otinef ide

10、ntifier serie?Da te speaficatonWork-file names optional)WF；Page：Cancel步驟：(3) ObjectNew ObjectOKCancelFile Edit Object | View PFOC Quick Options Add-ins Window Help New5crorte Seri Martisgje Links & Formulae-.s. fetchfrom OB,.Updat&-telect&d from DE.Store selected toVi 也Show Fetch Stor-e DeleteL era/

11、 oaiactarlRailSAr Rerame selected.F2Delete selected0巳 ri nt Selected步驟：(4丿 Type of objectPoolf j Workfile： New CJ bjectI 訕內(nèi)Frcr (Jfiji rr* nf objectRdnye. 1990ample 199CPqqIEqiwllc nc卜曰匚Tor Eldg dpllGroupLcALMatrisf-V ectcr -CZoefModelSampleScalar Series Sei i-es LifAk Series Alpha SAiOol SSwce SUlu

12、u SVcctor System Tabte- Text VsIMsp VARv Un titled步驟：（5）輸入所有序列名稱CE Pool: POOLMODEI Workfile; UNT|TLED;Untitled 口| View： | Proc | Object: 11 Print | Ma m e | Freeze j Estimate Define | PoolGenr Sheet Cross ticDkldTLti i rs：1 ers Ibalowthi s Um)AHBJFJHBHLJLJSJXLMNMGSDSHSXTJZJ步驟：(6 )定義各變量點(diǎn)擊sheet 輸入con

13、sume ? in come ? p?E Pool; POOLMODEL Workfile; UNTITLED;Untitled n xViewProcObject |PrintNameFreezeE 5timateDefinePoolGtnrCr*zs Seetion IdentLAM口FJHBHLJLJSJXLNNMGSDSHsxTJZJ步驟：（7）將表1、2、3中的數(shù)據(jù)復(fù)制到Eviews中Ot) sCONSUME?INCOME?p?CONSUME?INCOME?P?rAH-19563607.4304512.770109 9000 HAhA9Q73693.5504599.270101.30

14、00AH-19S83777.4104770 470100 DODOAH-19993901.310506460097.60000AH-2QC04232.9305293.5501007000AH-20C14517.6505569.8001Q0.5000AH-20C24736 5206032 40099.00000OJ-19965729 5207332 010111 6000BJ-19976531 8107813.160105.3000 0OJ-19986970.8308471 9801024000BJ-19Q97498.4809182.760100 5000BJ-2000049349010349

15、69103 50002估計(jì)操作:步驟:(1 )點(diǎn)擊 poolmodelEstimateV.jgw Pt. ：iiP rilL Miamiie PP ilGwrtr 5、1疋花1:l.Jxv n O i JniLk L AJtL fjttalf *X w Llhl. h 丄丄 Irx ,JAHPool Lstiilrlatiori燈 rm HU JLSHT iZTS pc cl fi cationO|pticinAOepe-j-ident griqbA* .： XA/oighlt 宅 2-IReArcs-soirzs ai-id ARQeil imEnni co ic-ffi ci e nl：s

16、 iFAtlFinaAJanpiTtlcq 打 I*.Mot KadiL Loiail冇己A.R.JSample：1956 2002No weightsCe larkce SdmpVv對(duì)話框說(shuō)明Pool EstiBationOptionsDtpn iRethdF ixed and RandLomCr2s_sic tiWeights No veightsEsiiirittiom sei iingxdrthod: LSL*ast Squares (and AR)Depe nden t variable:被解釋變量；Com mon :系數(shù)相同部分Cross-section specific:截面系數(shù)

17、不同部分步驟：（2）將截距項(xiàng)選擇區(qū)選Fixed effects （固定效應(yīng)）Cross-section : FixedPool EstimationSpeaficado n OptionsDepe ndent wadab亡 consume?Estirrajoft nietiod i(jJFixed and Rardcm EffectsRftg ； KSorsand ARQ termingCcmmoin coefflcfents!c in come?Crosi- ection 5pecific coefficient?：Period:V/eights: No weightsPeriod spec

18、ific coefficients!I IBalanceSamplesettingsMethod! L Least Squares (and AR)Saimpl?； 1596 2002得到如下輸出結(jié)果:D&p end ant Variable: CONSUME?Method. Pa&l&d Least SquaresDate: 07/16/14 Time: 11:06Sample: 1Q96 2002 included observations:1Cross-sections included: 15Total pool c&alaneed) observations: 105Variable

19、Coefficie ntStd Errort-statisticProbC50&504939645046.6302630.0000INCOME?0.B992320.01305049&4 8B20.0000Fijcsd Effects Cross)AH-C-5323597BJ-C592.4337FJ-C417S884HB-C-169.6295HLJC-192.03&4JL-C0.49391 &JAC-36 60391JX-C-341.5000LN-C76902hMG-C-230J 840SD-C-144321BSH-C327.1 OSOsx-c-95.13130TJ-C61.43642ZJ-C2

20、30.1580匚他 cte SpecifieatjonCross-section fixed Adummy vsrisblesjR-squared0992490M&an dependentvar4981.017Adjusted R. squared0.99122SS D dependentvar1700.995S.E. af regr&ssiori1593436Akaikft infb Hiterion13.11944Sumi squared resid2259742Schwart criteri on13.523B5Log likelihood-&727706HannarrQuinn ent

21、er.1328332FStatistic784.1521Durbin-Watson stat1.624146ProbfF-statistic0.000000接下來(lái)用F統(tǒng)計(jì)量檢驗(yàn)是應(yīng)該建立混合回歸模型，還是個(gè)體固定效應(yīng)回歸模Ho： i。模型中不同個(gè)體的截距相同（真實(shí)模型為混合回歸模型）。Hi：模型中不同個(gè)體的截距項(xiàng)i不同（真實(shí)模型為個(gè)體固定效應(yīng)回歸模型）對(duì)模型進(jìn)行檢驗(yàn)：(RRSS URSS)(4965275-2259743 )爪 1 =7 69 Fo( 14,90 )=18023URSS2259743/(NT N K1)790所以推翻原假設(shè)，建立個(gè)體固定效應(yīng)回歸模型更合理RRSS求法請(qǐng)參見(jiàn)Evi

22、ew面板數(shù)據(jù)之混合回歸模型相應(yīng)的表達(dá)式為：Co nsumet 596.50 0.69I ncomeit 53.230 592.44 D2. 230.16Di52(6.64)(49.55)R 0.99, SSEr 2259743其中虛擬變量D1,D2,., 65的定義是：1,如果屬于第個(gè)個(gè)體，i 12.,15 Di 0,其他15個(gè)省級(jí)地區(qū)的城鎮(zhèn)人均指出平均占收入68.62%。從上面的結(jié)果可以看出北京市居民的自發(fā)性消費(fèi)明顯高于其他地區(qū)。2.時(shí)點(diǎn)固定效應(yīng)模型時(shí)點(diǎn)固定效應(yīng)模型就是對(duì)于不同的截面（時(shí)點(diǎn)）有不同截距的模型。如果確知對(duì)于不同的截面，模型的截距顯著不同，但是對(duì)于不同的時(shí)間序列（個(gè)體）截距是相同

23、的，那么應(yīng)該建 立時(shí)點(diǎn)固定效應(yīng)模型：（2）（2），將時(shí)間項(xiàng)選擇KMt tkxkit uitk2時(shí)點(diǎn)固定效應(yīng)模型與個(gè)體固定效應(yīng)模型的操作區(qū)別在于步驟 區(qū)選Period : Fixed （時(shí)間固定效應(yīng)）Pool EstimaticnOp*mie|Uez in, nt 3 ar 卅 eAjcCTEEEorH nd *R ( tsnTE-11(irac incomer0 7245690 01057268.538140.0000Random Effects JCrcss)AH-C-2.553433BJ-C367 0439FJ-C-51.24006HB-C-104.8357HLJ-C-101 7600JL

24、-C54.90671JS-C-32.27363 j心-223.9519LN-C112.1152NMG-C-133.13775D-C-100.8713SH-AC126.1820XC*22,79189TJ-C100S794ZJ-C106 0939相應(yīng)的表達(dá)式是:cP? 345.2 0.72IRt2.6Di 367.0D2 . IO6.ID15（68.5）R2 0.98, SSE 2979246其中虛擬變量Di,D2,., D15的定義是：1,如果屬于笫i個(gè)個(gè)體,i=1,2,.,150,其他接下來(lái)利用Hausman統(tǒng)計(jì)量檢驗(yàn)應(yīng)該建立個(gè)體隨機(jī)效應(yīng)回歸模型還是個(gè)體固定效應(yīng)回歸模 型。H。：個(gè)體效應(yīng)與回歸

25、變量（IRt）無(wú)矢（個(gè)體隨機(jī)效應(yīng)回歸模型）Hi :個(gè)體效應(yīng)與回歸變量（IR ）相尖（個(gè)體固定效應(yīng)回歸模型）分析過(guò)程如下:得到如下檢驗(yàn)結(jié)果:Correlated Randan Effects * Hausman TestPool: POOLU2Test drosssection random effectsTest SummaryChiSq Statistic ChiSq. d.f. Prob.Cross-ceciiori random14.7875631 OIOOOICross-section random effects test I Dniparisoris:VariableFi 聽(tīng) d

26、Random Var (DifT.) Prob.IP?0.6975S107245690.000049010001由檢驗(yàn)輸出結(jié)果的上半部分可以看出，Hausman統(tǒng)計(jì)量的值是14.79，相對(duì)應(yīng)的概率是0.0001，即拒接原假設(shè)，應(yīng)該建立個(gè)體固定效應(yīng)模型。檢驗(yàn)結(jié)果的下半部分是Hausman檢驗(yàn)中間結(jié)果比較。個(gè)體固定效應(yīng)模型對(duì)參數(shù) 的估計(jì)值為 0.697561，隨機(jī)效應(yīng)模型對(duì)參數(shù)的估計(jì)值為0.724569。兩個(gè)參數(shù)的 估計(jì)量的分布方差的差為 0.000049。綜上分析，1996 -2002年中國(guó)東北、華北、華東15個(gè)省級(jí)地區(qū)的居民家庭人均消費(fèi)和人金收 入問(wèn)題應(yīng)該建立個(gè)體固定效應(yīng)回歸模型。人均消費(fèi)平均

27、占人均收入的70%隨地區(qū)不同，自發(fā)消費(fèi)(截距項(xiàng))存在顯著性差異。(4 )面板單位根檢驗(yàn)以cp序列為例。首先在工作文件窗口中打開(kāi)cp變量的15個(gè)數(shù)據(jù)組。ShowObj ects io display in I )d.i fferendi fferentInsliide in tAst equiticn .*) Iivdi i.MrO Indi I/dual ir snd and m.l)Nor 語(yǔ)OptiGTiSI Use balaiLcAd s5npl*Lug LengthSchwarzM AKImuni*_ 丿 4 JC* - indiant omat ics elec ti oil j m

28、i ax 1 mujn 1 ags jTVJTser sp&cii fiSyectral esti omICernelBartlettBandwidth selection(5 A-ntom a+ic :U”serereye s V得到如下檢驗(yàn)結(jié)果：GrouiJ unitoM 巳孚L一 SurcrrinrYiDate: 07/O2/QS Time* 15:40Sample: 1996 20C2Series: CPAHpCPBJ. CPFJ. CPHB, CPHU. CPJL. CPJS, CPJX, CP LN, CPNMG. CPSD, CP 呂比 CPSX. CFTJ. CPZJExoge

29、nous variables: Individual effectsAutomatic select ion of rnaxirrum PgsAutomatic select ion of lags based on SIC: 0 to 1Newey-West bandwidth selection using BArtlett kernelCross*MlethadStalistic Prab * eectioncObeNtir Uriii root (msEumEm 匚 omni or unit root process)Levin, Lin & Chu F9.6S7731,0000156

30、4Null： Unit root (assumes in dividual unit root process)Inn, Pesaran grid Shin W-slart6.463801.00001584ADF Fisher Chi-square3.049971 .ooao1584PF - Fisher Chi-square5 243351.00001590NliII: Nd unit root (Assunneg comirion unit yn process)Hadri 2-st at7.563830.000015105Probabilities for Fishar tests ar

31、e computed using 總 n asympotic Chi q uAre distribution. All other tests assume asymptotic normality.從上面的檢驗(yàn)結(jié)果可以看出來(lái)，6種檢驗(yàn)方法的結(jié)論都認(rèn)為位根。15個(gè)cp序列存在單選擇IPS檢驗(yàn)方法進(jìn)行單位根檢驗(yàn)。檢驗(yàn)結(jié)果如下:Null Hypothesis: Unit romt (iindividual unit root process)Date: 07702/06 Time： 15:44Sample: 1996 2002Series: CPAH.CPBJ, CPFJ. CPHB, CPHU.

32、 CPJL. CPUS. CPJXb CPLN.CPMMG. CPSD, CPSH, CPSX. CPTJ. CPZJEwogenous en ables： In dividual effectsAutomatic selection of nnaMimjin lagsAutomatic selection of lags based on SIC: 0 to 1Total number nf nbservalion 二 84Cross-sections in eluded1 15MethodStatisticProb.*Im, Pesaran and Shin W-stat6.463801.

33、0000* Probabilities are computed assuming asyrnpotic ncrmalityIntermediate ADF test resultsMaxSeriest-StstProb.E(t) Ety訥 L 朗Lac|bsCPAH5.55420.9999 -1.5562.G4S1 15CPBJ0.56130.9E95 -1.6472.3320 1GCPFJ-0 696907711-1.5472 3320 1GCPHB-0.53040 0149 -1.5472.3320 16CPHU2.G5S90.9991 -1.5472.3320 1GCPJL0.6324

34、0.9732 -1.6472干卩0 1GCPJS-1.36320.5305 -1.5472.3320 1GCPJX0.53560.96815472.3320 16CPLM1.127E0.9093 -1 5472.3320 1GCPNMG3.07E50.9992 /5 SB2 E481 15CPSD-0.33090.8450 -1.5582.6481 15CPSH0.79710.9754 -1.5562.6481 15CPSA2.11120.9900 -1.5472.3320 1GCPTJ-0.15460 88272.6481 15CPZJ2.00920 9966 -1 65B2.6481 1A

35、verage1 0653-1.5522.459從上面的結(jié)果可以看出 cp面板存在單位根，同時(shí)每個(gè)個(gè)體都存在單位根。2收集中國(guó)2000 -2005年各地區(qū)城鎮(zhèn)居民人均可支配收入X和消費(fèi)指出丫統(tǒng)計(jì)數(shù)據(jù)如表9.4。數(shù)據(jù)是6年的，每一年都有32組數(shù)據(jù)，共192組觀測(cè)值 人均可支配收入和消費(fèi)支出數(shù)據(jù)（單位：元）2 2可支可支配收可支可支可支配收可支地配收消費(fèi)消費(fèi)配收消費(fèi)配收消費(fèi)消費(fèi)配收消費(fèi)區(qū)入支出入支出入支出入支岀入支出入支出全X丫X丫X丫X丫X丫X丫6279.49986859.53097702.60298472.65109421.7182104937942國(guó)98.0058.0180.8820.9461

36、.10.03.88北103498493115778922124631028138821112156371220176521324京.2.924.60.623.84.840.40.954.20天8140.61218958.69879337.7191103127867114678802126389653津50.0470.225.26河北5661.43485984.44796679.50697239.54397951.58199107.669916.4782.7568.2806.7731.1809.67III4724.39415391.412

37、36234.47107005.51057902.56548913.6342西內(nèi)11.8705.0136.9603.3886.1591.635129.39275535.41956051.48597012.54198122.62199136.6928古05.7589.6200.8890.1499.2679.60遼5357.43565797.46546524.53427240.60778007.65439107.7369寧79.0601.4252.6458.9256.2855.27古4810.40205340.43376260.49737005.54927840.60688690.6794林黑00.

38、8746.2216.8817.1061.9962.71龍4912.38245425.41926100.44626678.50157470.55678272.6178江88.4487.3656.0890.1971.5351.01上117188868128839336132491046148671104166821263186451377海.00.804.00.490.34.821.03.033.41江6800.53237375.55328177.60429262.6708104817332123188621蘇23.1810.7464.606.57.82浙92

39、79.7020104647952117158713131799712145461063162931225江9.60.086.14.773.74安5293.42325668.45176032.47366778.50647511.57118470.6367徽55.9880.6540.5203.3443.3368.67福7432.56388313.60159189.66319999.7356111758161123218794建26.7408.1136.685.31.41江5103.36235506.38946335.45496901.491

40、47559.53378619.6109西58.5602.5164.3242.5564.8466.39Lb6489.50227101.52527614.55968399.60699437.6673107447457東97.0008.4136.3291.351河4766.38305267.41106245.45046926.49417704.52948667.6038南26.7142.1740.6812.6090.1997.02湖5524.46445855.48046788.56087321.59638022.63988785.6736北54.5098.7952.9298.25

41、75.5294.56湖6218.52186780.55466958.55747674.60828617.68849523.7504南73.7956.2256.7220.6248.6197.99廣9761.8016104158099111378988123809636136271069147691180東3.27.654.79.949.87廣5834.48526665.52247315.54137785.57638689.64459286.7032西43.3173.7332.4404.5099.7370.80海5358.40825838.43676822.5

42、4597259.55027735.58028123.5928南32.5684.8572.6425.4378.4094.79重6275.55696721.58737238.63608093.71189220.7973102438623慶98.8409.6904.2467.0696.05.46.29四5894.48556360.51766610.54137041.57597709.63718385.6891川27.7847.1780.0887.2187.1496.27貴5122.42785451.42735944.45986569.49487322.54948151.6159州21.2891.90

43、08.2823.9805.4513.29云6324.51856797.52527240.58277643.60238870.68379265.6996南64.3171.6056.9257.5688.0190.90西7426.55547869.59948079.69528765.80459106.83389431.8617藏32.4216.3912.4445.3407.2118.11陜5124.42765483.46376330.53786806.56667492.62338272.6656西24.6773.7484.0435.5447.0702.46甘4916.41265382.4420615

44、1.50646657.52987376.59378086.6529肅25.4791.3144.2424.9174.3082.20n5169.41855853.46986170.50426745.54007319.57588057.6245海96.7372.5952.5232.2467.9585.26寧4912.42005544.45956067.51046530.53307217.58218093.6404夏40.5017.4044.9248.3487.3864.31新5644.44226395.49316899.56367173.55407503.57737990.6207疆86.9304.

45、4064.4054.6142.6215.52首先建立工作文件，打開(kāi)工作文件后，過(guò)程如下: Torkfile: U1TTTILED1.Viewlpfoc I匝國(guó) 冋 nt lS3?e|Detaik+刀slete jGenr 1Ran ge:Sample： Ec0怕Few Obj ect. ” fienerat e VariesLinkslet ch fran DB Update slc-tQd. from DS ,Store selected tc DE.cnpy昶1弋吐畝EensneIDelete select elFrint Selected/-NewPagf/建立面板數(shù)據(jù)庫(kù)，并命名為XYD

46、isplay Filter. *電 for objectType of objectPoolEquationGrfliphGrouporCoe:Model SampleSer iesSeries LinkSeries AlphaSSpSystemWTftX tVsdMapVAJ輸入不同省市（包括全國(guó)）的標(biāo)識(shí)，如下:3 File Edit Object View Proc Quick Options Window Help object (Print “初韻帕62J Estimate (Defin PooiGenrsheet Cross Section Identifiers: (Enter i

47、dentifiers below this line) OOQGJIBJ32TJ33HB34SXD5NMG36LN37JLJ8HU39SH1OJS11ZJ12 AH13FJ14JX15SD16HN17HUB18HUN19GD胃F = unit tied2QQXPth = c: Kdocu/nnts &nd stingsi:h*uyi*iihyexits DB = none點(diǎn)擊sheet鍵，定義變量X和丫Eiews Pool: EY orkfxle: UNTITLEDr:Unt illedlFile Edi t Objact Vi aw Proc 蟲(chóng) lidc OAti OEU Vindow K

48、tlpuie 丹 Prcicl 3biec : I Prnt Ncme I |F eeze |Estimate DeLine I |F oc iGenr(5lee： |Cross Section Idtntififirs: (Enttr identifiers bslow this line)ODQGU1EJC2TJ 03HB H4SX 06NMGD6LIM07JL 09HU 09SH IOJS 11ZJ12AH13FJ14JX15SD1BHN 17HUE1SHUN19GD20GXFath = c : Wd0cumnt5 and xettingxVzhuyixi Wmy do-cuffinti

49、, DB = ncn WF = nmti tiedl_點(diǎn)擊Edit+/后，在數(shù)據(jù)窗口鍵入數(shù)據(jù)即可。對(duì)模型進(jìn)行估計(jì)，建立個(gè)體固定效應(yīng)回歸模型，過(guò)程如下:EYievs Pool: XT Vorkfile; UNTITLED:zUntitledJView Proc QuicM is. Window Help4998obsbjcct PrkitjhlamQ Freeze Edi 田-b如卜 B 呻 1+/_1 .mi Title | 戸比葉舊 1：亡Deh n 司卩 対 enrS3IHDI0Pool Estisation2迄2如 2EXZ.2E26XZ-200Specification OptionsDependent vari ahieRcercssors and AK 0 terms Common2J亓族刃27SHX-20I 27SHX-20I27S似郵 2Q 26GS-20C 28GS-20CEstimati on. mvtkodlFix 電d.Crcss5ecti ixed rrz KSi raPeri odNone2 X?Cross-section specificPeriod sptci ic28GS-2T28G3如匸29QHJ:2_得到如下輸出結(jié)果:Dependent Varnable： Y?