Scribd
Upload
84 views

DF-GLS vs. Augmented Dickey-Fuller: Elliott, Rothenberg, and Stock 1996

The document compares the DF-GLS unit root test to the Augmented Dickey-Fuller test. The DF-GLS test performs a modified Dickey-Fuller test on data that has been transformed via generalized least squares regression. It has greater power to detect stationarity compared to the Augmented Dickey-Fuller test. An example uses German macroeconomic data on log investment and finds that the DF-GLS test rejects the null hypothesis of a unit root, while the Augmented Dickey-Fuller test results are less conclusive.

Uploaded by

Shahabuddin Prodhan
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
84 views

DF-GLS vs. Augmented Dickey-Fuller: Elliott, Rothenberg, and Stock 1996

The document compares the DF-GLS unit root test to the Augmented Dickey-Fuller test. The DF-GLS test performs a modified Dickey-Fuller test on data that has been transformed via generalized least squares regression. It has greater power to detect stationarity compared to the Augmented Dickey-Fuller test. An example uses German macroeconomic data on log investment and finds that the DF-GLS test rejects the null hypothesis of a unit root, while the Augmented Dickey-Fuller test results are less conclusive.

Uploaded by

Shahabuddin Prodhan
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

DF-GLS vs.

Augmented Dickey-Fuller
This is almost completely taken from the Stata 11 Manual—Time-Series.
Title

[TS] dfgls DF-GLS unit-root test

Syntax

dfgls varname [if] [in] [, options]


options description

Main
maxlag( #) use # as the highest lag order for Dickey-Fuller GLS regressions
notrend series is stationary around a mean instead of around a linear time trend
ers present interpolated critical values from Elliot, Rothenberg, and Stock

dfgls tests for a unit root in a time series. It performs the modified Dickey–Fuller t test (known as the
DF-GLS test) proposed by Elliott, Rothenberg, and Stock (1996). Essentially, the test is an augmented
Dickey–Fuller test, similar to the test performed by Stata’s dfuller command, except that the time series is
transformed via a generalized least squares (GLS) regression before performing the test. Elliott,
Rothenberg, and Stock and later studies have shown that this test has significantly greater power than the
previous versions of the augmented Dickey–Fuller test.

dfglsperforms the DF-GLS test for the series of models that include 1 to k lags of the first differenced,
detrended variable, where k can be set by the user or by the method described in Schwert (1989). Stock
and Watson (Introduction to Econometrics, 2nd ed. 2007, 650–655) provide an excellent discussion of the
approach. As discussed in [TS] dfuller, the augmented Dickey–Fuller test involves fitting a regression of
the form

∆yt = α + βyt −1 + δt + ζ1∆yt −1 + ... + ζ k ∆yt − k + ut

and then testing the null hypothesis H0: β =0. The DF-GLS test is performed analogously but on GLS-
detrended data. The null hypothesis of the test is that yt is a random walk, possibly with drift. There are
two possible alternative hypotheses: yt is stationary about a linear time trend or yt is stationary with a
possibly nonzero mean but with no linear time trend. The default is to use the former. To specify the latter
alternative, use the notrend option.

References
Cheung, Y.-W., and K. S. Lai. 1995. Lag order and critical values of a modified Dickey–Fuller test. Oxford Bulletin
of Economics and Statistics 57: 411–419.

Elliott, G., T. J. Rothenberg, and J. H. Stock. 1996. Efficient tests for an autoregressive unit root. Econometrica 64:
813–836.

Ng, S., and P. Perron. 1995. Unit root tests in ARMA models with data-dependent methods for the selection of the
truncation lag. Journal of the American Statistical Association 90: 268–281.

-- . 2000. Lag length selection and the construction of unit root tests with good size and power. Econometrica 69:
1519–1554.

Schwert, G. W. 1989. Tests for unit roots: A Monte Carlo investigation. Journal of Business and Economic Statistics
2: 147–159.
Example
German macroeconomic dataset and test whether the natural log of income exhibits a unit root. Here
is the time series plot, which clearly exhibits a trend:
7
6.5
log investment
6
5.5
5

1960q1 1965q1 1970q1 1975q1 1980q1 1985q1


quarter

Try the augmented Dickey-Fuller test. Include a trend.


. dfuller ln_inv, lag(4) trend

Augmented Dickey-Fuller test for unit root Number of obs = 87

Interpolated Dickey-Fuller
Test 1% Critical 5% Critical 10% Critical
Statistic Value Value Value

Z(t) -3.133 -4.069 -3.463 -3.158

MacKinnon approximate p-value for Z(t) = 0.0987

The results from dfuller also support rejecting the null hypothesis, though the p-value with four lags is
close to 0.10. The problem with the ADF test is low power against the stationary alternative (especially
when you include a trend). So, we try the dfgls test.
Remember that the dfgls includes a trend by default.
. dfgls ln_inv

DF-GLS for ln_inv Number of obs = 80


Maxlag = 11 chosen by Schwert criterion

DF-GLS tau 1% Critical 5% Critical 10% Critical


[lags] Test Statistic Value Value Value

11 -2.925 -3.610 -2.763 -2.489


10 -2.671 -3.610 -2.798 -2.523
9 -2.766 -3.610 -2.832 -2.555
8 -3.259 -3.610 -2.865 -2.587
7 -3.536 -3.610 -2.898 -2.617
6 -3.115 -3.610 -2.929 -2.646
5 -3.054 -3.610 -2.958 -2.674
4 -3.016 -3.610 -2.986 -2.699
3 -2.071 -3.610 -3.012 -2.723
2 -1.675 -3.610 -3.035 -2.744
1 -1.752 -3.610 -3.055 -2.762

Opt Lag (Ng-Perron seq t) = 7 with RMSE .0388771


Min SC = -6.169137 at lag 4 with RMSE .0398949
Min MAIC = -6.136371 at lag 1 with RMSE .0440319

A maximum lag of 11 is selected by the Schwert criterion. The result suggests that 7 is optimal via the Ng-Perron
criterion. SC chooses 4 and MAIC only 1. This is a little confusing to me since the SC supposedly has a larger
penalty. I’d suggest taking a look at what the MAIC actually is. The null of a unit root is rejected at the 5%
level for lags 4–8 and 11 and at the 10% level for lags 9 and 10.

You might also like