Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: Same regression but different years | Diff in Diff

From   Julian Kochan <>
Subject   st: Same regression but different years | Diff in Diff
Date   Sun, 6 Apr 2014 00:42:14 +0200

Dear all,

I am using a panel data set. 2000 firms with observations from year 2000 to 2011. 
I am working with a fixed effect regression using the difference in difference approach. Dependent variable is the change in corporate investments. Independent variables are the year dummy post_2007, a treatment dummy, an interaction term and a few more other variables such as size, Tobin's Q, which are not important for my problem here.  
I created the post_2007 dummy such that post_2007 = 0 if year < 2007 and post_2007 = 1 otherwise. 
By creating such a variable and determine an interaction term = treatment*post:2007 I could investigate the effect of the financial crisis by comparing the period 2000-2007 with the period after the crisis 2007 till 2012. 

Now I am interested in investigating the change of corporate investments in other periods, in order to check the validity of my model in non crisis periods. 
I want now to check for the change from 2000 to 2001, and from 2001 to 2002, 2003 to 2004, and so on. I just want to compare the change between to years and then create the interaction term in order to see how investments have changed (for example) from 2000 to 2001. 

I just created dummies for each year. Is this the right approach ? But then, how will I create the interaction term ? The coefficient of the interaction term is the one of interest here. 
I would appreciate any help. Thank you so much

*   For searches and help try:

© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index