Panel data model,

For econometric discussions not necessarily related to EViews.

Moderators: EViews Moderator, EViews Gareth

Panel data model,

Postby canpelican on Sat Aug 11, 2012 1:31 pm

Hello everybody,

I am quite new to eViews, I'm using SPSS usually, but this time, it cannot help me anymore, so, I switched to EViews.
Here is the deal: I have a panel data, which after a great amount of time and nerve consumption in the process of collecting it, is supposed to be analyzed. It is also the first time I'm dealing with panel data analysis, so everything is new for me. And to be even better, the result of the analysis should be inserted as soon as possible in my graduation project (which makes it a problem of life or death, graduate or not).

The model I want to analyze is this:

Cost of capital = b0 +b1Early + b2Late + b3 Early*Mandatory + b4 Late*Mandatory +b5 Mandatory + ∑〖bi 〗Company Specific Variables + ∑〖bi 〗Country Specific Variables

Where:
b0 = intercept

Early, Late and Mandatory are binary (dummy) variables, which define companies in one of this groups: Early, Late, Mandatory.
Early* Mandatory and Late*Mandatory are Interaction Terms .

Company specific variables are (with expected sign of influence):
(-) ROA (return on assets)
(+) Leverage
(-) logTotalAssets
(+) Bid-Ask Spread


Country specific variables (with expected sign of influence) :
(+) Inflation
(-) Index of economic globalization


So, my data is collected from a few number of countries, each company is measured for a number of 7 years, and is placed in one of the 3 categories: Early, Late, Mandatory.
My panel data are balanced, which means that the number of companies * 7 years = an integer number of observation (the cross-sections). However, I still have some data missing but I think this should not be a problem.

Can someone tell me the standard procedure for panel data analysis? For my equation, what are the steps to be done?
Mainly, I'm interested to find how is my dependent variable behave in time, for each group of companies (Early, Late, Mandatory).
My theoretical expectation is that the dependent variable is supposed to decrease, and this decrease is expected to be greater for Early group of companies, then for Late and lastly for Mandatory. The country variables are supposed to control for the economic background of the country, and this might have an impact on the output.

Thank you very much!!!
canpelican
 
Posts: 1
Joined: Sat Aug 11, 2012 12:01 pm

Re: Panel data model,

Postby ayoolafemi1 on Fri Sep 07, 2012 1:35 am

Dear ALL,
Kindly, safe me. can anybody help me i want to estimate static panel data model in the presence of heteroscedasticity. I need the written codes.
Thank u in advance
ayoolafemi1
 
Posts: 1
Joined: Fri Sep 07, 2012 1:19 am


Return to Econometric Discussions

Who is online

Users browsing this forum: No registered users and 4 guests