2
$\begingroup$

Has anyone used Excel or R Studio to do a simple Chow Test (compare a set of data before and after one particular year)? I have tried to find info but there is not much, especially regarding the use of Excel.

I have monthly CDS values from January 2004 to December 2010. I have to split them in two - 2004-2006 and 2007-2010 - and I have to see how they changed from one period to another. The values are in basis points.

$\endgroup$
4
  • $\begingroup$ Isn't this just done in Excel by calculating a linear regression for each period, and calculating the Chow statistic using the squared residuals from each regression? Otherwise the R package strucchange should do it. $\endgroup$ Commented Aug 1, 2016 at 13:28
  • $\begingroup$ As I am quite new to regression methods I want to be sure that I understand what I am doing so please bear with me just a little! Now, this is what I do not get: Which of the Excel functions I must use for the Chow Test? Thank you! $\endgroup$
    – RobertC
    Commented Aug 1, 2016 at 13:41
  • $\begingroup$ The Chow test helps us figure out if two sets of data may be pooled because they have the same slope and intercept. Could you take a look at iuj.ac.jp/faculty/kucc625/method/panel/chow_test.pdf (and an example in Stata) and maybe ask your question again? $\endgroup$
    – rajah9
    Commented Aug 2, 2016 at 13:41
  • $\begingroup$ @RobertC: A very warm welcome to Quant.SE and thank you for your question. See my answer below: quant.stackexchange.com/a/28452/12 $\endgroup$
    – vonjd
    Commented Aug 3, 2016 at 7:35

1 Answer 1

2
$\begingroup$

You can do a Chow test with the strucchange package in R. You can find the full documentation here.

The package comes with a vignette which shows how to use it (see p. 9ff. if you want to dive into an example without further ado).

I don't know whether there is an appropriate Excel add-in available but in general I would not recommend doing these more sophisticated analyses in Excel anyway.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.