Two-way fixed effects model

[Paola Vivoli]

Hi Statalisters,

I have a panel data which spans from 2008 through 2015 and covers 181 Italian listed family firms. My main interest is the relation between founding-family ownership and firm performance. The analysis also incorporates variables that identify CEOs as firm founders, descendants of the firm's founder, or outsiders. I would like to use a two-way fixed effects model for my regression analysis.

The paper I have read that does something similar describes the fixed effects to be dummy variables for each year of the sample and dummy variables for each two-digit SIC code (I would like to use ATECO 2007 Code since I am talking about Italy), and the regression they employ is the following:

Firm Performance= δ0 + δ1 (Family Firm) + δ3 (control Variables) + δ3 + δ54 (Two digit ATECO Code) + δ'93-'99 (Year Dummy Variables) + 𝛆

where
Firm Performance = ROA based on EBITDA and net income, and Tobin's q;
Family Firm = binary variable that equals one when the founding family is pre- sent in the firm, and zero otherwise; Control Variables = officer and director holdings less family holdings, fraction of independent directors serving on the board, research and development expenses divided by total sales, long-term debt divided by total as- sets, stock return volatility, natural log of total assets, and the natural log of firm age;
Two-Digit ATECO Code = 1.0 for each two-digit SIC code in our sample;
Year Dummy Variables = 1.0 for each year of our sample period."

How should I build the model on STATA?

Thank you a lot!

[Carlo Lazzaro]

Paola:
you may want to take a look at -xtreg,fe- to start off.
Please, see the FAQ on how to post more effectively and about reposting the same query when you got no reply (your second try is probably better than the first one, though. The main downside rests on the fact that you're inadvertently appearing like asking somebody else to do this part of the research on your behalf).

[Paola Vivoli]

Thank you Carlo,

The thing is that I tried off kinda everything, and there are different and contrasting opinions re. two-way fixed effects models and since I am new to STATA I have lots of doubts.
From what I have understood, -xtreg, fe simply does not takes into account the two-way fixed effects model, and I am not sure the suggestion by my Professor to create the new variable

-egen id = group(Year ATECO)

and then running an OLS regression that replicates the two-way fixed effects really does the job, namely:

-areg ROA FamilyFirm Officersanddirectorsownless DE DEBITDA Returnperemployee Salesgrowth CapexPPE1 LTDTA1 Returnvolatility1 lnTotalassets Lnfirmage, absorb (id)

[Carlo Lazzaro]

Paola:
thanks for providing further details efficiently.
You might be interested in this quite old Stata thread (please note that -xi- notation is now unnecessary): https://www.stata.com/statalist/arch.../msg00730.html
Besides, googling with the string -two-way fixed effect Stata- gives back several promising entries (such as http://myweb.fsu.edu/tsass/Papers/Re...Estimation.pdf).

[Paola Vivoli]

Thank you a lot Carlo! I don't know anyone who knows how to properly use STATA so really thank you for your help!
I will now read everything carefully.

I have in the meantime tried out some different models and I have seen that the results are the same for two of them and quite similar to the third model (following the suggestion of my professor), so everything seems to work quite right. Namely:

- reg ROA FamilyFirm Officersanddirectorsownless CapexPPE1 DE DEBITDA LTDTA1 Returnvolatility1 lnTotalassets Lnfirmage i.ATECO i.Year

- areg ROA FamilyFirm Officersanddirectorsownless CapexPPE1 DE DEBITDA LTDTA1 Returnvolatility1 lnTotalassets Lnfirmage i.Year, absorb(ATECO)

- areg ROA FamilyFirm Officersanddirectorsownless CapexPPE1 DE DEBITDA LTDTA1 Returnvolatility1 lnTotalassets Lnfirmage, absorb(id)


I will now double check with what you sent!

Thanks again

[Paola Vivoli]

Carlo, again thank you a lot! I compared the 2 more models from the first link with my 3 models and they all give the same results (but the one the Professor suggested me to do)!
I will then check with him but I finally solved my doubts.

Thanks a lot again and I wish you a very nice day.

Best,
Paola

[Carlo Lazzaro]

Paola:
glad with reading that you've made it.
I do reciprocate the very same to you.