could any one suggest me how to use the Zandrews command to know stationary properties of a variable? I am getting the result at level for intercept, trend and both. As i am getting the result only at level, could anyone suggest me the estimation command for first difference?
Hello!
You can try the following Stata command to run Zivot and Andrews (1992) unit root test:
zandrews D.varname, lagmethod(AIC) break(both)
zandrews D.varname, lagmethod(AIC) break(intercept)
zandrews D.varname, lagmethod(AIC) break(trend)
Hello!
You can find the STATA output as follows:
"Allowing for break in intercept
Lag selection via AIC: lags of D.D.gnp96 included = 1
Minimum t-statistic -6.176 at 1979q1 (obs 49)
Critical values: 1%: -5.34 5%: -4.80 10%: -4.58"
To interpret this:
The null hypothesis for Zivot-Andrews Unit Root Test is:
the time series has unit root with structural break in the intercept.
This null hypothesis can be rejected when the critical values of (1%, 5% and 10%) are greater than the test statistics value.
- Badges
- Science topic
- Similar topics
- Statistics
- Time Series