Hello, I'm trying to run a regression to test whether firms’ capital structure changes more during the 2007-2008 financial crisis according to their pre-crisis liquidity. The crisis period is from 2007Q3 - 2008Q2. My quarterly data runs from 2000 - 2020 and I've created a dummy variable for the crisis period.

For the pre-crisis liquidity, I'm using dummy variable to indicate LOW and HIGH liquidity level based on industry median

My baseline model is something like this: LEVit = PRE-LIQUID + CRISIS + PRE-LIQUID x CRISIS + CONTROLS (1)

In the second model, I want to use real data for the PRE-LIQUID instead of dummy variable. So I created PRE-LIQUID and pasted only the liquidity value of 2007Q2 (the quarter before the crisis) to this variable and ran the same regression model (1).

I'm not sure if this is the right way to carry out the test. Also, about the timeframe, will it be better if I restrict the period to only 2006 - 2010 instead?

Any advice is appreciate. Thank you so much!