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
Hi all,
Vanilla are planning an update to the site on April 24th (next Wednesday). It is a major PHP8 update which is expected to boost performance across the site. The site will be down from 7pm and it is expected to take about an hour to complete. We appreciate your patience during the update.
Thanks all.

Regression help

  • 03-04-2014 12:55pm
    #1
    Helpneeded86
    Registered Users Posts: 523 ✭✭✭


    I have to do a regression as part of my thesis but I am pretty much close to new with them. Can anyone point a good place to start?

    How basic or advanced will the regression be?

    My dependent variable is share price movement for a 3 year period taken at year end at each year.

    My independent variable is female directors as a % of the board.


Comments

  • #2
    Iwasfrozen
    Registered Users Posts: 16,250 ✭✭✭✭Iwasfrozen


    Helpneeded86 wrote: »
    I have to do a regression as part of my thesis but I am pretty much close to new with them. Can anyone point a good place to start?

    How basic or advanced will the regression be?

    My dependent variable is share price movement for a 3 year period taken at year end at each year.

    My independent variable is female directors as a % of the board.
    Stata would be the most common econometrics software package, your university should provide you access free of charge.


  • #3
    Helpneeded86
    Registered Users Posts: 523 ✭✭✭Helpneeded86


    I have done the regression. I would just like confirmation that the results pretty much mean my data does not show any real relation.

    302281.jpg


  • #4
    Iwasfrozen
    Registered Users Posts: 16,250 ✭✭✭✭Iwasfrozen


    Your R^2 is pretty low which would indicate a small level of relation. But like anything in statistics it really depends on where you want to set the cut off point.


  • #5
    andrew
    Moderators, Science, Health & Environment Moderators, Society & Culture Moderators Posts: 3,368 Mod ✭✭✭✭andrew


    OP, do you have any other independent variables?


  • #6
    Helpneeded86
    Registered Users Posts: 523 ✭✭✭Helpneeded86


    andrew wrote: »
    OP, do you have any other independent variables?

    I updated it with board size. These are the results I got.

    303652.jpg


  • Advertisement
  • #7
    andrew
    Moderators, Science, Health & Environment Moderators, Society & Culture Moderators Posts: 3,368 Mod ✭✭✭✭andrew


    Do you have any other independent variables which might also explain a company's share price movement? As in, what you want to show is that the %of female directors is significant, after taking other relevant factors into account. So for example, share prices are plausibly affected whether there's an ongoing recession or not. So you might want to control for that. There are likely many papers which try to use regressions to explain share price movements using a variety of factors, so you could try searching google scholar for some ideas as to what variables to include.

    Also, re. the dependent variable. You said it's the share price movement. As in, are you looking to see whether more women = less volatile share price? Or a more volatile share price? (i.e. are you looking at the variance in the share price). Or are you looking to see whether more women means share price growth/or contraction, (i.e. the growth rate of the share price over the period).

    Also, make sure that you've got a solid theory as to why the number of women on a company's board might affect share price at all. I presume you've thought about that, but I figured it was worth reiterating; a nice regression result doesn't prove anything in and of itself.


  • #8
    Helpneeded86
    Registered Users Posts: 523 ✭✭✭Helpneeded86


    andrew wrote: »
    Do you have any other independent variables which might also explain a company's share price movement? As in, what you want to show is that the %of female directors is significant, after taking other relevant factors into account. So for example, share prices are plausibly affected whether there's an ongoing recession or not. So you might want to control for that. There are likely many papers which try to use regressions to explain share price movements using a variety of factors, so you could try searching google scholar for some ideas as to what variables to include.

    Also, re. the dependent variable. You said it's the share price movement. As in, are you looking to see whether more women = less volatile share price? Or a more volatile share price? (i.e. are you looking at the variance in the share price). Or are you looking to see whether more women means share price growth/or contraction, (i.e. the growth rate of the share price over the period).

    Also, make sure that you've got a solid theory as to why the number of women on a company's board might affect share price at all. I presume you've thought about that, but I figured it was worth reiterating; a nice regression result doesn't prove anything in and of itself.

    I am looking for a relation between share price and the number of female directors. There has been a lot of studies and most used ROA and ROE. My model is going to be a lot more simplistic.

    I decided to minimise the regression end of this research as I am simply not very good at it. I explored other avenues in the area in order to make the project look good. I understand the regression is below par compared to what it could be but I gues it is best to work with your strengths.


Advertisement