Advertisement
If you have a new account but are having problems posting or verifying your account, please email us on hello@boards.ie for help. Thanks :)
Hello all! Please ensure that you are posting a new thread or question in the appropriate forum. The Feedback forum is overwhelmed with questions that are having to be moved elsewhere. If you need help to verify your account contact hello@boards.ie

Econometrics/ regression help!

Options
  • 19-11-2011 12:15pm
    #1
    lolliemc
    Registered Users Posts: 21 ✭


    I'm currently trying to run regressions on oxmetrics to find out if corporate governance in companies is related to shareholders' wealth. The variables I have collected from my data are: shareholders' funds, total compensation of CEO, bonus of CEO, salary of CEO, if CEO is also chairman and the number of independent directors. I can't seem to get my regressions to work. What variables should I be testing against shareholders funds in order to find my results?

    Thanks!!!:)


Comments

  • Options
    #2
    mari2222
    Registered Users Posts: 568 ✭✭✭mari2222


    unrelated - why should they be related?. Except perhaps in family owned firms. My opinion only.


  • Options
    #3
    TiGeR KiNgS
    Registered Users Posts: 1,435 ✭✭✭TiGeR KiNgS


    mari2222 wrote: »
    unrelated - why should they be related?. Except perhaps in family owned firms. My opinion only.

    I think OP is trying to find if there is a premium on shares for good corporate governance practices.


  • Options
    #4
    andrew
    Moderators, Science, Health & Environment Moderators, Society & Culture Moderators Posts: 3,368 Mod ✭✭✭✭andrew


    lolliemc wrote: »
    I'm currently trying to run regressions on oxmetrics to find out if corporate governance in companies is related to shareholders' wealth. The variables I have collected from my data are: shareholders' funds, total compensation of CEO, bonus of CEO, salary of CEO, if CEO is also chairman and the number of independent directors. I can't seem to get my regressions to work. What variables should I be testing against shareholders funds in order to find my results?

    Thanks!!!:)

    I assume you want to know if corporate governance affects shareholder wealth, right? So your dependent variable (Y) would be shareholder wealth. The explanatory variable of interest is corporate governance right? So you need to have some sort of measure of corporate governance in your regression; this is crucial. If you don't have data for the explanatory variable, they you can't test your hypothesis! Then throw in the other variables you mentioned above as further explanatory variables, and run the regression. It'll look something like

    y(shareholder value) = b1(measure of corporate governance) + b2(CEO comp) + b3(bonus) + b4(salary) + b5(is chairman) + b6(indie directors)


    Use all the variables at first, and you can remove the ones which aren't significant by using an F test (I think). Don't forget to check for multicollinearity and heteroskedasticity. Also, there might be some reverse causation going on, with shareholder value determining CEO compensation as well as the other way around.


  • Options
    #5
    SBWife
    Registered Users Posts: 1,287 ✭✭✭SBWife


    There's going to be autocorrelation in the equation given above as compensation will be by definition be related to CEO salary and bonus. I'd run at least two separate regressions one including compensation as the independent variable and another including the two components of that compensation ie. salary and bonus.

    The other aspect you don't seem to be including which is considered an important measure of NED independence and corporate governance is the length of tenure of the NEDs.


Advertisement