Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, is already up and running.

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Re: st: xtreg individual and year fixed effects

From   Christopher Baum <>
To   "" <>
Subject   Re: Re: st: xtreg individual and year fixed effects
Date   Sun, 1 Apr 2012 07:14:15 -0400

On Apr 1, 2012, at 2:33 AM, Daniele wrote:

> So now you suggest me to generate 2 new variables, one that represents
> the country-product fe and one that represents the year fe and then to
> sum them to the residual.
> I think that something like this should work:
> by countryproduct: egen cp_fe=mean(ife)
> by year: egen year_fe=mean(ife)
> gen newvar= cp_fe + year_fe + resid
> Thank you very much
> Daniele
> 2012/3/29 Christopher Baum <>:
>> <>
>> webuse grunfeld,clear
>> xtreg invest mvalue kstock i.year, fe
>> predict double ife, u
>> replace ife = ife + _b[_cons]
>> forv i=1936/1954 {
>> Â  Â  Â  Â qui replace ife = ife + _b[`i'.year] if year==`i'
>> }
>> su ife
>> tabstat ife, by(year)
>> tabstat ife, by(company)

No, it will not work. If you do the first three commands with grunfeld and then do

tabstat ife, by(company) 
tabstat ife, by(year)

you will see that ife has a (approx) zero mean for every year, by construction. That is because it is the individual fixed effect. To take account of the year fixed effects you need to do EXACTLY what I have spelled out above. 

Furthermore, why would you want to add the FE to the residual? I thought you were trying to identify the non-idiosyncratic piece.
If you want the individual FE + residual, then predict option -ue- will give you that. Then all you need to do is add the year effect as above.


Kit Baum   |   Boston College Economics & DIW Berlin   |
                             An Introduction to Stata Programming  |
  An Introduction to Modern Econometrics Using Stata  |

*   For searches and help try:

© Copyright 1996–2015 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index