matlab计量经济学工具箱Word下载.docx

matlab计量经济学工具箱Word下载.docx

《matlab计量经济学工具箱Word下载.docx》由会员分享，可在线阅读，更多相关《matlab计量经济学工具箱Word下载.docx（22页珍藏版）》请在冰豆网上搜索。

Amodelobjectholdsalltheinformationnecessarytoestimate,simulate,andforecasteconometricmodels.

Parametricformofthemodel

Numberofmodelparameters（e.g.,thedegreeofthemodel）

Innovationdistribution（GaussianorStudent'

st）

Amountofpresampledataneededtoinitializethemodel

Example1:

AR

（2）

wheretheinnovationsareindependentandidenticallydistributednormalrandomvariableswithmean0andvariance0.2.Thisisaconditionalmeanmodel,sousearima.

>

model=arima（'

AR'

{0.8,-0.2},'

Variance'

0.2,'

Constant'

0）

Example2:

GARCH（1,1）model

model=garch（'

GARCH'

NaN,'

ARCH'

NaN）

或者

model=garch（1,1）

ParameterswithNaNvaluesneedtobeestimatedorotherwisespecifiedbeforeyoucanforecastorsimulatethemodel.

TodisplaythevalueofthepropertyARforthecreatedvariableobject，

model.AR

model.Distribution=struct（'

Name'

'

t'

DoF'

8）

Methodsarefunctionsthatacceptmodelobjectsasinputs.InEconometricsToolbox,

estimate

infer

forecast

simulate

Example3:

FitanARMA（2,1）modeltosimulateddata

1）Simulate500datapointsfromtheARMA（2,1）model

simModel=arima（'

{0.5,-0.3},'

MA'

0,'

0.1）;

rng（5）;

Y=simulate（simModel,500）;

2）SpecifyanARMA（2,1）modelwithnoconstantandunknowncoefficientsandvariance.

model=arima（2,0,1）;

model.Constant=0

3）FittheARMA（2,1）modeltoY.

fit=estimate（model,Y）

Example4:

loadData_EquityIdx

nasdaq=Dataset.NASDAQ;

r=price2ret（nasdaq）;

r0=r（1:

2）;

rn=r（3:

end）;

FitaGARCH（1,1）modeltothereturns,andinfertheloglikelihoodobjectivefunctionvalue.

model1=garch（1,1）;

fit1=estimate（model1,rn,'

E0'

r0）;

[~,LogL1]=infer（fit1,rn,'

Wold'

stheorem:

youcanwriteallweaklystationarystochasticprocessesinthegenerallinearform

Thus,byWold'

stheorem,youcanmodel（orcloselyapproximate）everystationarystochasticprocessas

Theconditionalmeanandvariancemodels

StationaritytestsIfyourdataisnotstationary,considertransformingyourdata.Stationarityisthefoundationofmanytimeseriesmodels.

Youcandifferenceaserieswithaunitrootuntilitisstationary,Or,considerusinganonstationaryARIMAmodelifthereisevidenceofaunitrootinyourdata.

SeasonalARIMAmodelsuseseasonaldifferencingtoremoveseasonaleffects.Youcanalsoincludeseasonallagstomodelseasonalautocorrelation.

ConductaLjung-BoxQ-testtotestautocorrelationsatseverallagsjointly.Ifautocorrelationispresent,considerusingaconditionalmeanmodel.

Lookingforautocorrelationinthesquaredresidualseriesisonewaytodetectconditional

Heteroscedasticity.Tomodelconditionalheteroscedasticity,considerusingaconditionalvariancemodel.

YoucanuseaStudent’stdistributiontomodelfattertailsthanaGaussiandistribution（excess

kurtosis）.

Youcancomparenestedmodelsusingmisspecificationtests,suchasthelikelihoodratiotest,Wald’stest,orLagrangemultipliertest.

TheJohansenandEngle-Grangercointegrationtestsassessevidenceofcointegration.ConsiderusingtheVECmodelformodelingmultivariate,cointegratedseries.Itcanintroducespuriousregressioneffects.

Theexample“SpecifyingStaticTimeSeriesModels”explorescointegrationinstaticregression

models.Type>

showdemoDemo_StaticModels.

WhyTransform?

Isolatetemporalcomponentsofinterest.

Removetheeffectofnuisancecomponents（likeseasonality）.

Makeaseriesstationary.

Reducespuriousregressioneffects.

Stabilizevariabilitythatgrowswiththeleveloftheseries.

Maketwoormoretimeseriesmoredirectlycomparable.

P207

Anexampleofastaticconditionalmeanmodelistheordinarylinearregressionmodel.

Adynamicconditionalmeanmodelspecifiestheevolutionoftheconditionalmean,

Examples:

ByWold’sdecomposition,youcanwritetheconditionalmeanofanystationaryprocessytas

And

istheconstantunconditionalmeanofthestationaryprocess.

arima（p,D,q）:

nonseasonalARterms（p）,theorderofnonseasonalintegration（D）,andthenumberofnonseasonalMAterms（q）.

Whensimulatingtimeseriesmodels,onedraw（or,real