Augmented Dickey-Fuller Trend and Intercept

Question

I am trying to determine if I should include an "intercept" or a "trend and intercept" when using the Augmented Dickey-Fuller (ADF) test. I ran a regression with my dependent variable and a time trend and both the intercept and trend were statistically significant. Is this statistical significance what determines if I should include a "trend and intercept" in the ADF test? My data is unstructured/undated. The series contains 252 observations and there are about 6 observations per year from 1970-2014. Below are my regression results and a graph of the raw data.

enter image description here

Have you checked out existing posts on Dickey-Fuller test? See, for example, [here](http://stats.stackexchange.com/questions/44647/which-dickey-fuller-test-should-i-apply-to-a-time-series-with-an-underlying-mode) and [here](http://stats.stackexchange.com/questions/8351/trend-or-no-trend). — Richard Hardy, Mar 03 '15 at 07:21

score -1 · Answer 1 · edited Jan 29 '16 at 08:09

-1

library(tseries)
library(urca)
data2<-na.remove(data1$SANDP) #assumed one dataset
    data2
    c<-adf.test(data2,k=2)
    c
    summary(c$statistic)

useful command for you

summary(ur.df(data2,type=c("drift"),lags=2,selectlags =c("Fixed")))

edited Jan 29 '16 at 08:09

Learner

1,528
3
18
34

answered Jan 29 '16 at 06:46

Harshil Gandhi

1
1

you can change type=drift /none/trend. – Harshil Gandhi Jan 29 '16 at 06:47

Augmented Dickey-Fuller Trend and Intercept

1 Answers1