0
$\begingroup$

So, for my research paper I have to write an equation for my multiple regression model. My independent variable is STOCK PRICE; my independent variable is before/after (denoted TIME on my equation) which means it is a dummy variable with the values of 0 and 1; I also have two control variables of PROFIT and CAPITALIZATION.

So what I have usually seen when building a model is two subscripts i and j, where i is the time and j is the company. However the same company in my sample can appear more than once given that the before/after variable refers to specific events which can occur more than once in a company. Therefore, should I add a third subscript for the event?

My final formula would look like this

PRICEijk = β1TIMEijk + β2PROFITijk + β3SIZEijk

Would this be alright or is there some rule that regressions should only have the two subscripts i and j?

Or would it be better to define one subscript in terms of two parameters? i.e. subscript i refers to firm X impacted by event Y, while k refers to the time period?

Improve this question
$\endgroup$
2
  • $\begingroup$ From the design you refer, I think you should do an interaction term between time and company, which will then tell you if there is any variability within time across companies (I assume this may be a variable of interest for you). $\endgroup$
    – André Barros
    Commented Aug 29, 2022 at 18:44
  • $\begingroup$ "there some rule that regressions should only have the two subscripts i and j?" there are no rules for notation like this; simply write so as to be easily understood. If you think 3 or 4 or 10 indices is the clearest way to express your model, do that. $\endgroup$
    – John Madden
    Commented Aug 29, 2022 at 21:00

0

Reset to default

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.