Calculating large amounts of betas at once
Posted: Sun Jul 22, 2012 2:33 am
Dear all,
I am fairly new to Eviews and I've been stuck for a couple of days with the following problem.
I have slightly more than 4,000 annual returns for approximately 700 different companies over a time period of 11 years. Meaning that the time periods differ per company. I also have dummy variables for every company year, eight in total (the portfolios to which the companies belong).
Following the Fama MacBeth methodology, I want to test whether the factors explain the portfolio returns and the premia awarded to exposure to each factor.
The first step in this analysis is a regression between the return of a certain company year on the eight factors. And I need this for all 4000+ firm years. What I want in the end is a beta for every firm year (4000+ betas).
Do you have any idea how to generate these betas in an efficient way? As I said, I'm new to Eviews and I cant figure it out.
I am fairly new to Eviews and I've been stuck for a couple of days with the following problem.
I have slightly more than 4,000 annual returns for approximately 700 different companies over a time period of 11 years. Meaning that the time periods differ per company. I also have dummy variables for every company year, eight in total (the portfolios to which the companies belong).
Following the Fama MacBeth methodology, I want to test whether the factors explain the portfolio returns and the premia awarded to exposure to each factor.
The first step in this analysis is a regression between the return of a certain company year on the eight factors. And I need this for all 4000+ firm years. What I want in the end is a beta for every firm year (4000+ betas).
Do you have any idea how to generate these betas in an efficient way? As I said, I'm new to Eviews and I cant figure it out.