Page 1 of 1

Elliot- Rothenberg-Stock unit root test

Posted: Sat Dec 07, 2013 12:01 pm
by iliasvalasop
I was asked to to a more powerful unit root test on my data than ADF and Phillips-Perron.
Elliot- Rothenberg-Stock unit root test was the one pointed out and I was reading carefully the User Guide2 but i didn't find out what my results mean, because I am not familiar to the test.
this is the table:

Null Hypothesis: LOGGDP has a unit root
Exogenous: Constant
Lag length: 1 (Spectral OLS AR based on SIC, maxlag=10)
Sample: 1948 2014
Included observations: 67

P-Statistic

Elliott-Rothenberg-Stock test statistic 1603.793
Test critical values: 1% level 1.897200
5% level 3.017600
10% level 3.998400

*Elliott-Rothenberg-Stock (1996, Table 1)

HAC corrected variance (Spectral OLS autoregression) 0.000986

Is ther a unit root or not??

Thank you so much for your time once again

Re: Elliot- Rothenberg-Stock unit root test

Posted: Fri Dec 19, 2014 5:29 am
by playboy
Why there is nobody answers this question?

Re: Elliot- Rothenberg-Stock unit root test

Posted: Fri Dec 19, 2014 6:55 am
by trubador
Reasons of not getting an answer may vary depending on the question. For this particular one, however, the answer is (was) already in the output so there is nothing to say other than a simple Yes or No. Since null hypothesis, test statistic and critical values are all you need to decide whether a series has a unit root, I prefer user(s) to figure out the answer on their own for the sake of learning process.