空间面板数据分析R的splm包.docx
- 文档编号:24807008
- 上传时间:2023-06-01
- 格式:DOCX
- 页数:14
- 大小:3.24MB
空间面板数据分析R的splm包.docx
《空间面板数据分析R的splm包.docx》由会员分享,可在线阅读,更多相关《空间面板数据分析R的splm包.docx(14页珍藏版)》请在冰豆网上搜索。
空间面板数据分析R的splm包
空间面板数据分析——R的splm包(总11页)
空间面板数据分析——R的splm包
(任建辉,暨南大学)
ThesplmpackageprovidesmethodsforfittingspatialpaneldatabymaximumlikelihoodandGM.
安装R软件及其编辑器Rstudio
网址:
www.r-project.org
下载好Rstudio以后,操作都可以Rstudio中完成了,包括命令的编写、命令运行、图形展示,最方便的要数查看数据了。
R界面
Rstudio界面,形如matlab
下面进入正题,了解splm包中的数据、命令及结果展示。
所有命令都写在编辑窗口(studio左上区域),可以单独的运行每行命令,也可选取一段一起执行,点run按钮。
1、首先,安装splm包并导入,命令如下:
intall.packages(“splm”),选择最近的下载点
library(splm)
>library(splm)
载入需要的程辑包:
MASS
载入需要的程辑包:
nlme
载入需要的程辑包:
spdep
载入需要的程辑包:
sp
载入需要的程辑包:
Matrix
载入需要的程辑包:
plm
载入需要的程辑包:
bdsmatrix
载入程辑包:
‘bdsmatrix’
下列对象被屏蔽了from‘package:
base’:
backsolve
载入需要的程辑包:
Formula
载入需要的程辑包:
sandwich
载入需要的程辑包:
zoo
载入程辑包:
‘zoo’
下列对象被屏蔽了from‘package:
base’:
as.Date,as.Date.numeric
载入需要的程辑包:
spam
载入需要的程辑包:
grid
Spamversion0.40-0(2013-09-11)isloaded.
Type'help(Spam)'or'demo(spam)'forashortintroduction
andoverviewofthispackage.
Helpforindividualfunctionsisalsoobtainedbyaddingthe
suffix'.spam'tothefunctionname,e.g.'help(chol.spam)'.
载入程辑包:
‘spam’
下列对象被屏蔽了from‘package:
bdsmatrix’:
backsolve
下列对象被屏蔽了from‘package:
base’:
backsolve,forwardsolve
载入需要的程辑包:
ibdreg
载入需要的程辑包:
car
载入需要的程辑包:
lmtest
载入需要的程辑包:
Ecdat
载入程辑包:
‘Ecdat’
下列对象被屏蔽了from‘package:
car’:
Mroz
下列对象被屏蔽了from‘package:
nlme’:
Gasoline
下列对象被屏蔽了from‘package:
MASS’:
SP500
下列对象被屏蔽了from‘package:
datasets’:
Orange
载入需要的程辑包:
maxLik
载入需要的程辑包:
miscTools
Pleasecitethe'maxLik'packageas:
Henningsen,ArneandToomet,Ott(2011).maxLik:
ApackageformaximumlikelihoodestimationinR.ComputationalStatistics26(3),443-458.DOI10.1007/s00180-010-0217-1.
Ifyouhavequestions,suggestions,orcommentsregardingthe'maxLik'package,pleaseuseaforumor'tracker'atmaxLik'sR-Forgesite:
https:
//r-forge.r-project.org/projects/maxlik/
Warningmessage:
程辑包‘Matrix’是用R版本3.0.3来建造的
注意:
在导入splm时,如果发现还有其他配套的包没有安装,需要先安装。
2、接着,查看数据及结构,命令如下:
data(Produc,package=”Ecdat”)
View(Produc)
3、引入空间权重矩阵(spatialweightsmatrix),命令如下
data(usaww)
Views(usaww)
4、空间面板数据的广义矩估计,命令spgm
GM<-spgm(log(gsp)~log(pcap)+log(pc)+log(emp)+unemp,data=Produc,
listw=usaww,moments=”fullweights”,spatial.error=TRUE)
summary(GM)
>GM<-spgm(log(gsp)~log(pcap)+log(pc)+log(emp)+unemp,data=Produc,
+listw=usaww,moments="fullweights",spatial.error=TRUE)
>summary(GM)
SpatialpanelfixedeffectsGMmodel
Call:
spgm(formula=log(gsp)~log(pcap)+log(pc)+log(emp)+unemp,
data=Produc,listw=usaww,spatial.error=TRUE,moments="fullweights")
Residuals:
Min.1stQu.Median3rdQu.Max.
-0.14000-0.01950-0.003160.015300.16800
Estimatedspatialcoefficient,variancecomponentsandtheta:
Estimate
rho0.3277625
sigma^2_v0.0012179
Coefficients:
EstimateStd.Errort-valuePr(>|t|)
log(pcap)-0.00224350.0262646-0.08540.9319295
log(pc)0.24149790.023582610.2405<2.2e-16***
log(emp)0.78132760.028385527.5256<2.2e-16***
unemp-0.00360260.0010094-3.56910.0003582***
---
Signif.codes:
0‘***’0.001‘**’0.01‘*’0.05‘.’0.1‘’1
5、空间面板极大似然估计,命令spml
fm<-log(gsp)~log(pcap)+log(pc)+log(emp)+unemp
##fixedeffectspanelwithspatialerrors
Fespaterr<-spml(fm,data=Produc,listw=mat2listw(usaww),model=”within”,
spatial.error=”b”,hess=FALSE)
summary(Fespaterr)
>fm<-log(gsp)~log(pcap)+log(pc)+log(emp)+unemp
>Fespaterr<-spml(fm,data=Produc,listw=mat2listw(usaww),model="within",
+spatial.error="b",hess=FALSE)
>summary(Fespaterr)
Spatialpanelfixedeffectserrormodel
Call:
spml(formula=fm,data=Produc,listw=mat2listw(usaww),model="within",
spatial.error="b",hess=FALSE)
Residuals:
Min.1stQu.Median3rdQu.Max.
-0.1250-0.0238-0.00350.01710.1880
Coefficients:
EstimateStd.Errort-valuePr(>|t|)
rho0.55740130.032955416.9138<2e-16***
log(pcap)0.00514380.02507240.20520.83745
log(pc)0.20530260.02319968.8494<2e-16***
log(emp)0.78225400.027874128.0638<2e-16***
unemp-0.00223170.0010735-2.07880.03764*
---
Signif.codes:
0‘***’0.001‘**’0.01‘*’0.05‘.’0.1‘’1
##randomeffectspanalwithspatiallag
Respatlag<-spml(fm,data=Produc,listw=mat2listw(usaww),model=”random”,
spatial.error=”none”,lag=TRUE)
summary(Respatlag)
>Respatlag<-spml(fm,data=Produc,listw=mat2listw(usaww),model="random",
+spatial.error="none",lag=TRUE)
>summary(Respatlag)
SpatialpanelrandomeffectsMLmodel
Call:
spreml(formula=formula,data=data,index=index,w=listw2mat(listw),
w2=listw2mat(listw2),lag=lag,errors=errors,cl=cl)
Residuals:
Min.1stQu.MedianMean3rdQu.Max.
1.381.571.701.701.802.13
Errorvarianceparameters:
EstimateStd.Errort-valuePr(>|t|)
phi21.31758.30172.56780.01023*
Spatialautoregressivecoefficient:
EstimateStd.Errort-valuePr(>|t|)
lambda0.1616150.0290995.5542.793e-08***
Coefficients:
EstimateStd.Errort-valuePr(>|t|)
(Intercept)1.658149950.1507185511.0016<2.2e-16***
log(pcap)0.012945050.024939970.51900.6037
log(pc)0.225553760.0216342210.4258<2.2e-16***
log(emp)0.670810750.0264211325.3892<2.2e-16***
unemp-0.005797160.00089175-6.50097.984e-11***
---
Signif.codes:
0‘***’0.001‘**’0.01‘*’0.05‘.’0.1‘’1
6、伴有随机效应和序列误差相关的空间面板模型的极大似然估计,命令speml
##randomeffectspanelwithspatiallagandserialerrorcorrelation
##optimizationmethodsetto“BFGS“
Sarsrmod<-spreml(fm,data=Froduc,w=usaww,error=”sr”,lag=TRUE,method=”BFGS”)
summary(Sarsrmod)
>Sarsrmod<-spreml(fm,data=Produc,w=usaww,error="sr",lag=TRUE,method="BFGS")
>summary(Sarsrmod)
SpatialpanelrandomeffectsMLmodel
Call:
spreml(formula=fm,data=Produc,w=usaww,lag=TRUE,errors="sr",
method="BFGS")
Residuals:
Min.1stQu.MedianMean3rdQu.Max.
2.663.023.183.183.313.77
Errorvarianceparameters:
EstimateStd.Errort-valuePr(>|t|)
psi0.997263530.000821381214.1<2.2e-16***
Spatialautoregressivecoefficient:
EstimateStd.Errort-valuePr(>|t|)
lambda0.3029420.0303769.973<2.2e-16***
Coefficients:
EstimateStd.Errort-valuePr(>|t|)
(Intercept)1.236702930.227775545.42955.652e-08***
log(pcap)0.082579770.036173712.28290.02244*
log(pc)0.015099190.019773240.76360.44510
log(emp)0.738820210.0293414425.1801<2.2e-16***
unemp-0.002709620.00065851-4.11483.875e-05***
---
Signif.codes:
0‘***’0.001‘**’0.01‘*’0.05‘.’0.1‘’1
7、模型检验
(1)bsjktest,Baltigi,Song,Jung,andKohLMtestforspatialpanels
>bsjktest(fm,data=Produc,listw=usaww,test="C.1")
Baltagi,Song,JungandKohC.1conditionaltest
data:
log(gsp)~log(pcap)+log(pc)+log(emp)+unemp
LM=0.2617,df=1,p-value=0.609
alternativehypothesis:
spatialdependenceinerrorterms,subREandserialcorr.
(2)bsktest,Baltigi,SongandKohLMtestforspatialpanels
>bsktest(fm,data=Produc,listw=mat2listw(usaww),
+test="LM1",standardize=TRUE)
Baltagi,SongandKohSLM1marginaltest
data:
log(gsp)~log(pcap)+log(pc)+log(emp)+unemp
SLM1=0.083,p-value=0.9338
alternativehypothesis:
Randomeffects
(3)Covarianceextractormethodforsplmobjects
>sarremod<-spml(fm,data=Produc,listw=mat2listw(usaww),model="random",
+lag=TRUE,spatial.error="none")
>library(lmtest)
>coeftest(sarremod)
ztestofcoefficients:
EstimateStd.ErrorzvaluePr(>|z|)
(Intercept)1.658149950.1507185511.0016<2.2e-16***
log(pcap)0.012945050.024939970.51900.6037
log(pc)0.225553760.0216342210.4258<2.2e-16***
log(emp)0.670810750.0264211325.3892<2.2e-16***
unemp-0.005797160.00089175-6.50097.984e-11***
---
Signif.codes:
0‘***’0.001‘**’0.01‘*’0.05‘.’0.1‘’1
>library(car)
>lht(sarremod,"log(pcap)=log(pc)")
Linearhypothesistest
Hypothesis:
log(pcap)-log(pc)=0
Model1:
restrictedmodel
Model2:
function(x,...)
UseMethod("formula")
DfChisqPr(>Chisq)
1
2136.2681.719e-09***
---
Signif.codes:
0‘***’0.001‘**’0.01‘*’0.05‘.’0.1‘’1
(更多详情请查看splm的help文档以及文后列的参考文献)
参考文献
1.Baltagi,B.H.,Song,S.H.,JungB.andKoh,W.(2007)Testingpaneldataregressionmodelswithspatialandserialerrorcorrelation.JournalofEconometrics,140,5-51
2.Baltagi,B.H.,Song,S.HandKoh,W.(2003)Testingpaneldataregressionmodelswithspatialerrorcorrelation.JournalofEconometrics,117,123-150
3.Millo,G.,Piras,G.(2012)splm:
SpatialPanelDataModelsinR.JournalofStatisticalSoftware,47
(1),1-38.URLhttp:
//www.jstatsoft.org/v47/i01/
4.Elhorst,J.P(2003)Specificationandestimationofspatialpaneldatamodels,InternationalRegionalScienceReview,26,pages244-268
5.Elhorst,J.P(2009)Spatialpaneldatamodels,InFisher,M.M.andGetis,A.(eds),HandbookofAppliedSpatialAnalysisSpringer,Berlin
6.GiovanniMilloandGaetanoCarmeci,(2011)“Non-lifeinsuranceconsumptioninItaly:
asubregionalpaneldataanalysis”,JournalofGeographicalSystems,13:
273-298
7.QuFengandWilliamC.Horrace,(2012)”AlternativeMeasuresofTechnicalEfficiency:
Skew,BiasandScale”,JournalofAppliedEconometrics,Forthcoming.
8.Kapoor,M.,Kelejian,H.H.andPrucha,I.R.(2007)Paneldatamodelwithspatiallycorrelatederrorcomponents,JournalofEconometrics,140,pages97-130
9.Mutl,J.,andPfaffermayr,M.(2011)TheHausmantestinaCliffandOrdpanelmodel,EconometricsJournal,14,pages48-76
10.Kelejian,H.H.andPrucha,I.R.(1999)AGeneraliedMomentsEstimatorfortheAutoregressiveParameterinaSpatialModel,InternationalEconomicReviews,40,pages509-533
11.Kelejian,H.H.andPrucha,I.R.(1999)AGeneraliedSpatialTwoStageLeastSquareProcedureforEstimatingaSpatialAutoregressiveModelwithAutoregressiveDisturbances,JournalofRealEstateFinanceandEconomics,17,pages99-122
12.Millo,G.(2013)Maximumlikelihoodestimationofspatiallyandseriallycorrelatedpanelwithrandomeffects.
- 配套讲稿:
如PPT文件的首页显示word图标,表示该PPT已包含配套word讲稿。双击word图标可打开word文档。
- 特殊限制:
部分文档作品中含有的国旗、国徽等图片,仅作为作品整体效果示例展示,禁止商用。设计者仅对作品中独创性部分享有著作权。
- 关 键 词:
- 空间 面板 数据 分析 splm