Quantcast
Channel: Statalist
Viewing all articles
Browse latest Browse all 73267

Conflicting result in split sample and interaction sample

$
0
0
Hi, let me explain my problem. I am trying to examine the effect of indep variable A during the GFC, which i defined as between 2007-2008. So I tried two methods. My sample period is between 2000-2013

1) I split the sample into Pre (2000-2006), During (2007-2008) and Post (2009-2013) and run the regression "reg dep A i.year" And indep variable A is 0.1216*** (Pre), 0.0523 (during) and 0.00035 (post), with no significance during during and post periods.

2) I create a GFC dummy variable that equal to 1 if the Year is 2007-2008. And interact GFC*A, I run "reg dep A GFC*A io(2008 2009).year". My result for GFC is 0.0899*** and GFC*A is 0.0495***. This imply that, during GFC, there is an incremental effect on A. Note that because GFC is a dummy, I only controlled for year effect excluding the year dummies for 2007 and 2008.

However, if I look at method 1, during GFC, it wasnt significant, moreover, the coefficient reduces. Isnt the results supposed to tally in some way? Any advice is much appreciated.

Viewing all articles
Browse latest Browse all 73267

Trending Articles



<script src="https://jsc.adskeeper.com/r/s/rssing.com.1596347.js" async> </script>