Hi,

I wanted to know if someone can help me. I am trying to ascertain the change in the sensitivity of one of my variables to a change in another independent variable.

i have run the following regression : xtreg leadcompnew PrdmktFLUID ROE_w ln_sale_w LEV MTB VOLT MKTRET TENURE gender2 AGE c.PrdmktFLUID#c.ROE_w i.year if inrange(year,2012,2016), fe vce(cluster gvkey).

I want to know how the relationship between leadcomnew(y) and ROE_w(x) changes at different levels of PrdmktFLUID(so 25th and 75th percentile).

leadcompnew = compensation
ROE_w = performance proxy
PrdmktFLUID = competition proxy

I tried using this code: lincom B1*ROE +B*Fluidity at 25th percentile *ROE

which gave me this result:

lincom -1.1128*ROE_w +0.4621*1.486*ROE_w

( 1) - .4261194*ROE_w = 0

------------------------------------------------------------------------------
leadcompnew | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
(1) | .4741889 .2488671 1.91 0.057 -.0146535 .9630313

I am unsure of how to interpret this. I want to know the economic significance as well as the statistical significance.
Would my next step be to do one at 75th percentile and then just take the difference?

Thanks in advance