資源描述：
《協(xié)整檢驗(yàn)及誤差修正模型的建立》由會(huì)員上傳分享，免費(fèi)在線閱讀，更多相關(guān)內(nèi)容在行業(yè)資料-天天文庫。
1、協(xié)整檢驗(yàn)及誤差修正模型的建立 1978年-2003年度北京農(nóng)村居民人均純收入與人均消費(fèi)支出的統(tǒng)計(jì)表單位：（元/人）年份197819791980198119821983198419851986人均純收入R193.25191.01191.47221.34241.64238.22229.29239.05253.45人均消費(fèi)支出P184.89175.97155.07180.99197.75197.81186.63187.26210.6年份198719881989199019911992199319941995人均純收入R285.77343.543
2、86.02399.17408.09406.86486.01584.39661.74人均消費(fèi)支出P243.52290.85315.9356.92336.45329.9403.66504.15552.83年份19961997199819992000200120022003人均純收入R743.29811.19848.58904.13925.54964.711015.921046.33人均消費(fèi)支出P594.86593.65620.57626.11617.22633.6671.1692.73一．為了方便計(jì)算，首先將數(shù)據(jù)取對(duì)數(shù)化得到新的統(tǒng)計(jì)表：obs
3、LNPLNR19785.2197610536422615.26398468746753219795.170313525963675.25232578269704419805.0438766278760585.25473113838511419815.1984417811207815.39969998083663919825.2870036066477645.4874490152705419835.2873069740279845.47319461645275519845.2291280471950385.43498757792290219
4、855.2324980255107435.47667273507211219865.3499605996998755.53516656491835919875.4951990746733215.6551872914767419885.6728076703567715.8393035532088819895.755425707808045.95588918159406119905.8775116670319515.98938739131278219915.8184495496343266.01148773831344919925.79878
5、95782345366.00846914592380619936.0005729394719736.18622919982049519946.2228738428392246.3705685681627219956.315050540156056.4948727295461119966.3883260837419796.61108627812265419976.3862899201118016.69850230534476319986.4306384105300266.74356436426046419996.43952607451319
6、6.80697315536291920006.4252255243709946.83037735096074520016.4514178405002166.87182753803964820026.5089181571641946.92354988488077720036.5406403128824456.95304408244083二．檢驗(yàn)時(shí)間序列的平穩(wěn)性從圖中發(fā)現(xiàn)，LNP和LNR有大致相同的增長和變化趨勢，說明二者可能存在協(xié)整關(guān)系。此時(shí)間序列存在隨時(shí)間增長的趨勢，因此，模型應(yīng)存在時(shí)間項(xiàng)。三．對(duì)LNP進(jìn)行ADF檢驗(yàn)NullHypothe
7、sis:LNPhasaunitrootExogenous:Constant,LinearTrendLagLength:3(Fixed)t-Statistic??Prob.*AugmentedDickey-Fullerteststatistic-2.226120?0.4533Testcriticalvalues:1%level-4.4407395%level-3.63289610%level-3.254671從中我們看到t統(tǒng)計(jì)量是-2.2261,大于顯著性水平為1%(5%,10%)的臨界值,表明序列LnP是不平穩(wěn)的。四．對(duì)LNR進(jìn)行ADF檢
8、驗(yàn)NullHypothesis:LNRhasaunitrootExogenous:Constant,LinearTrendLagLength:3(Fixed)t-Statistic??Prob
