Estimating Earnings Persistence

[Laura Westhof]

Dear Stata experts,

For my thesis, I investigate earnings persistence by the following model:

where alpha1 measures the earnings persistence.

I have the data on the earnings and outstanding stocks. Is there a Stata command to estimate the alpha1 and subsequently use this estimation as a variable?

Thank you in advance!

[Andrew Musau]

You need to create the defined variable:

gen y= earnings/outstanding_stocks
xtset firm_id year
regress y L.y, cluster(firm)

\(\widehat{\alpha}_1\) is the estimated coefficient on L.y.

[Laura Westhof]

Thank you Andrew Musau for your quick response!

Your explanation is very clear. I have a follow-up question: How can I use this estimation of alpha 1 as a dependent variable? Because I would like to measure the effect of the independent variable on earnings persistence.

[Andrew Musau]

\(\widehat{\alpha}_1\) is a coefficient. Even if you include it as a variable, it is a constant. You should talk to your supervisor/ colleague who is familiar with what you are doing. Perhaps they want you to estimate a coefficient for each firm and then use the estimated coefficients as a dependent variable in a cross-sectional regression. I am only speculating here.

[Laura Westhof]

I will do that. Thank you for your response!