I am conducting a regression analysis estimating the impact of event-country credit ratings on the CDS spreads of non-event countries. As of now the current formula looks like this:
'pct_chg_-1_1 ~ Event'
I also have information about the credit rating prior to the new announcement for both the event country and non-event country and would like to incorporate this information into the regression model. For now I have transformed both credit ratings into a dummy variable where above investment grade ratings are equal to 1.
Hence, as of now my idea is the following:
'pct_chg_-1_1 ~ Event + inv_grade_event + inv_grade_non_event'
However, this doesn't account for the different possible combinations:
- Event and non-event are of investment grade
- Event and non-event are below investment grade
- Only the event country is of investment grade, non-event country is below
- Only the non-event country is of investment grade, event country is below
Is there a better way to incorporate this information?