1. weak stationarity


Source: Enders 2003:213 Example



Download 161.63 Kb.
Page4/4
Date23.04.2018
Size161.63 Kb.
#46690
1   2   3   4

Source: Enders 2003:213

Example

We would like to perform unit roots for the ln_VAT



Step1: Perform ADF test using the most general structure and record statistic



i.e VAT has a unit root (random walk with drift and time trend)

i.e. VAT is stationary

In Eviews this is very easy since the unit root options are sufficient. The initial selection lag order ,k, should be determined as i.e. integer of . The alternative is to use the automatic lag selection criteria such as the Bayesian information criterion, Swartz information criterion, Akaike information criterion



Go to view/unit root test

ADF Test Statistic

-0.349651

1% Critical Value*

-3.5572







5% Critical Value

-2.9167







10% Critical Value

-2.5958

*MacKinnon critical values for rejection of hypothesis of a unit root.































Augmented Dickey-Fuller Test Equation

Dependent Variable: D(LN_VAT)

Method: Least Squares

Date: 11/13/09 Time: 16:50

Sample(adjusted): 2002 2054

Included observations: 53 after adjusting endpoints

Variable

Coefficient

Std. Error

t-Statistic

Prob.

LN_VAT(-1)

-0.020222

0.057835

-0.349651

0.7281

D(LN_VAT(-1))

-0.493401

0.118000

-4.181357

0.0001

C

0.205849

0.489533

0.420501

0.6759

R-squared

0.282226

Mean dependent var

0.019825

Adjusted R-squared

0.253516

S.D. dependent var

0.193443

S.E. of regression

0.167134

Akaike info criterion

-0.685106

Sum squared resid

1.396684

Schwarz criterion

-0.573580

Log likelihood

21.15531

F-statistic

9.829928

Durbin-Watson stat

2.091018

Prob(F-statistic)

0.000251

Perform hypothesis test using the critical values from Eviews. Remember that the the ADF unit root test is left tail.

In our case we cannot reject the null of unit roots(Remember the ADF test is a left tail test).



In our case the computed ADF statistic is in the acceptance/fail to reject region. Thus we fail to reject the null hypothesis that money supply m3 has unit roots with drift and time trend.



Repeat the test

Download 161.63 Kb.

Share with your friends:
1   2   3   4




The database is protected by copyright ©ininet.org 2024
send message

    Main page