Page 1 of 1

ADF test basic interpretation

Posted: Mon May 09, 2011 4:29 am
by caruros
Hello all.
I am new to this forum and need your help as quickly as posible.
I am writing disertation on the topic of monetary and fiscal policy effectivness and have variables M3 and Goverment spending as representation of those politics. my model is
GDP= c + B1*M3+B2* goverment spending +u
I have several economterics and e views quesitons.

First of all i have tested data for stationarity by using ADF test. I ended up with following results:


ADF Test Statistic -1.645959 1% Critical Value* -3.6019
5% Critical Value -2.9358
10% Critical Value -2.6059

If i understood coreclty this means that i have non stationary data because my ADF statistic is lower then critical values?
When i test ADF with trend i get ADF HIGHER then critical value. Does this means data is stationary but with trend? how do i find the trend and use it in regresion then? Further, if one variabale is non stationary does it mean that i can difference only that variable, or i must difference both of them?
Please help its my disertation and i may fail if i do not fix this.