Dependent Variable on the firm level and independent variables on the country level

For econometric discussions not necessarily related to EViews.

Moderators: EViews Gareth, EViews Moderator

jazzhamlet
Posts: 1
Joined: Tue May 18, 2021 5:27 pm

Dependent Variable on the firm level and independent variables on the country level

Postby jazzhamlet » Wed May 19, 2021 3:17 am

Dear all,

I am working on a paper to find out how democracy in a certain country impact the growth of investments of multinational corporations. I did something similar on Foreign Direct Investment (country level) in that case it was a straight forward panel data analysis as my independent variables (Main or control ) were also on the country level (level of democracy, GDP, Inflation ... etc). However, now I am trying to find out how individual multinational corporation increase or decrease its investment in relation to changes in the level of democracy. For that I have data on firm level (1000 company) over 12 years time period, those companies invest in 56 countries. Therefore my dependent variable is on the firm level and independent variables on the country level.
My problem is the repeated data on the country level, For example GM invests in Brazil and China, and Ford also invests in Brazil and China. So when I insert the data there are repeated country level data such as GDP in Brazil will be repeated for both Ford and GM. Can I do that? It does not sound right to me. Also, am I dealing here with multidimensional Panel data here? Should I create a dummy variable for country? I am frankly quite lost, so any suggestions for me to start would really appreciated!

startz
Non-normality and collinearity are NOT problems!
Posts: 3700
Joined: Wed Sep 17, 2008 2:25 pm

Re: Dependent Variable on the firm level and independent variables on the country level

Postby startz » Wed May 19, 2021 7:30 am

Nothing wrong with what you're doing.

Country level fixed effects would probably be a good idea.

vikingpeggy
Posts: 1
Joined: Wed May 26, 2021 4:42 am

Re: Dependent Variable on the firm level and independent variables on the country level

Postby vikingpeggy » Thu May 27, 2021 11:56 am

I would go -logistic- or -logit- with standard errors clustered at -country- level.

https://www.downloadtexasgisdata.com/do ... ounty-gis/ | https://www.downloadtexasgisdata.com/do ... ounty-gis/


Return to “Econometric Discussions”

Who is online

Users browsing this forum: No registered users and 7 guests