@BREITUNG implements the test for unit roots in panel data from Breitung(2000). The Breitung test as a null of a unit root with an alternative of a common dominant stationary root. Alternative tests for unit roots in panel data are @LevinLin, @Hadri and @IPShin.

@BREITUNG( options )   series start end

Parameters

Options

DET=NONE/[CONSTANT]/TREND

Individual-specific deterministic components to remove. Note NONE is an error and is provided simply to keep the form of the option similar to related procedures.

 

SMPL=standard SMPL option [not used]

 

LAGS=(maximum) number of additional lags of the differenced series to include [Schwert's suggestion]

CRIT=FIXED/[GTOS]/AIC/BIC/HQ

Criterion used in selecting the individual specific lags. FIXED uses the value of the LAGS option for each. GTOS is General TO Specific, starting at the value of the LAGS option and dropping lags as long as the t-stat on the final one has a significance level below the SLSTAY option. AIC, BIC and HQ are the Akaike, Schwarz Bayesian and Hannan-Quinn, respectively.

SLSTAY=significance level for keeping the marginal lag in CRIT=GTOS [.10]

 

ROBUST/[NOROBUST]

Uses robust (Eicker-White) standard errors in computing the test statistic.

 

[PRINT]/NOPRINT

TITLE=title for output ["Breitung Unit Root Test Series: ..."]

Example

open data pedroni_ppp.xls

calendar(panelobs=246,m) 1973:6

data(format=xls,org=columns) 1//1973:06 20//1993:11 ae rf wpi cpi uswpi uscpi country entry year month

*

set logratio  = log(cpi/uscpi)

set logexrate = log(ae)

*

@breitung(det=trend,crit=gtos) logratio

 

Sample Output

Breitung Unit Root Test: Series LOGRATIO

Test has large N, large T (sequential) asymptotics

Null is Unit Root. Alternative is Common Stationary Root.

Individual Specific Components: Constant+Trend

With average lags 13.10 chosen from 15

 

N                    20

T                   246

Test Statistic -6.05409

Signif          0.00000

 

Copyright © 2025 Thomas A. Doan