Interpreting the constant when performing a fixed effect panel data regression in Stata

244 Views Asked by At

Dear Stackexchange community,

I am running a panel data regression on 20 years of monthly historical excess returns of the stocks in the S&P 500 at 31/12/2017. I like to test the effectiveness of several factor models, however I am having difficulty interpreting the constant that Stata provides. If I recall correctly, when using a fixed effect model the equation when using one independent variable is as follows.

$$Y_{i,t}=βX_{i,t}+α_i+u_{i,t}$$

where $\alpha_i$ is an individual intercept per stock.

When I perform a fixed effect panel data regression of my dataset on the monthly excess market return provided by Kenneth French on his website, stata gives me a constant. How should I interpret that? My regression result was as follows:

My regression then would follow this equation right?:

$$r_{i,t}-r[\text{risk free}]_t= β(r[mkt]_t-r[\text{risk free}]_t)+α_i+u_{i,t}$$

my regression result is the following: panel data regression result (xtreg, fe)

as you can see, it includes a constant. How do I interpret this?

any help would be greatly appreciated! kind regards,

Marinus