Basic Dickey Fuller Test
by theorangedog on Nov.02, 2007, under Skills
See the attached document for an example of how to perform a Dickey Fuller test for cointegration, along with an excerpt of the distinct t-table.
This example determines if a single data set is cointegrated. Replacing this series with the desired measure of spread between two series, or another metric such as the log ratio of two series, may be done as a method for identifying pairs trading candidates.
4 comments for this entry:




January 27th, 2008 on 12:55 pm
[...] of the most popular posts on my site to date concerns the Dickey Fuller test, and the most downloaded document is the Dickey Fuller [...]
February 10th, 2008 on 6:01 pm
[...] variable and the single independent variable, for a unit root. This can be done by using the Dickey-Fuller test. Outcome 1: Neither variable has a unit root. In this case, least squares can be used to estimate [...]
April 23rd, 2008 on 7:51 am
Thanks for this example, been looking for something simple and straightforward!
Question; In your example, you utilize 10% as the critical value based on your sample size. What does the 10% represent, confidence interval? I see in your note that Excel only reports the 10% critical value…Just looking for clarification.
Where does the basic DF test breakdown or become weak from a trading perspective? I see more references to the Johansen approach and Engle-Granger approach.
Thanks!!
February 4th, 2010 on 5:46 am
Hi,
I have some problems on how to perform a Dickey Fuller test, and really hoped that your Dickey Fuller Example could help me. The problem is that I can’t open the example file.
Is there something wrong with the link? If not, is it possible to get the example by e-mail.
Thanks,
Hanne