I would kindly like to ask for guidance regarding my question below.
To illustrate my question: I have two industries: Industry 1 and Industry 2. Both industries contain a number of firms that belong to that specific industry. When the stock return (%) of the firms in an industry is averaged, Industry 1 has a higher (average) return than Industry2.
Using firm-specific characteristics (leverage, coverage ratio, size etc) I would like to find out whether any of these variables explain the return difference between the two industries. My question is how to do that. Should I use a pooled regression and interaction dummies?
The literature uses various forms of regressions which confused me, therefore, I was wondering if anyone could help.
Thank you in advance.
You better conduct pre-estimation tests first to determine whether your data meet all the OLS assumptions. If one of the important assumptions is not fulfilled, eg. heteroscedasticity, multicolinearity, cross sectional dependence or autocorrelation, then you need to look for a more suitable model that fits the data.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Regression