[Attaullah Shah]

asreg reports both R2 and adjusted R2. See the following example

. webuse grunfeld, clear

. asreg invest mvalue kstock, by(company )

. ed

. sort year

. list _* in 1/10

     +--------------------------------------------------------------------+
     | _Nobs         _R2      _adjR2   _b_mvalue   _b_kstock      _b_cons |
     |--------------------------------------------------------------------|
  1. |    20   .68040764   .64280854   .16237772   .00310173    22.707115 |
  2. |    20   .70530671   .67063691   .02655119   .15169387   -9.9563082 |
  3. |    20   .92135403   .91210157   .11928082   .37144481    -149.7824 |
  4. |    20   .91357845   .90341121   .07794781   .31571818   -6.1899547 |
  5. |    20   .66551452   .62616329   .07538795   .08210356   -7.7228397 |
     |--------------------------------------------------------------------|
  6. |    20   .76350084   .73567741   .08752719   .12378141    -4.499533 |
  7. |    20    .7444461   .71438094   .05289412   .09240652   -.50938763 |
  8. |    20   .64315782   .60117638   .00457343    .4373692    .16151866 |
  9. |    20    .9521422   .94651187   .13145485   .08537425   -8.6855436 |
 10. |    20   .47086233   .40861084   .17485601   .38964188   -49.198313 |
     +--------------------------------------------------------------------+

.

[Kate Lussy]

Dear Professor Shah,

Thank you for your response. However, when I use, for example asreg F_Excess_USD IVOL, fmb newey(6), I only get the average R^2, just as below.

Fama-MacBeth Two-Step procedure (Newey SE) Number of obs = 51312
(Newey-West adj. Std. Err. using lags(6)) Num. time periods = 317
F( 1, 316) = 1.03
Prob > F = 0.3115
avg. R-squared = 0.0416
------------------------------------------------------------------------------
| Newey-FMB
F_Excess_U~w | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
IVOL_w | -.0612065 .0603753 -1.01 0.311 -.1799948 .0575818
_cons | .0086751 .0016891 5.14 0.000 .0053517 .0119985
------------------------------------------------------------------------------

In addition, tried to compute idiosyncratic volatility on a rolling basis of 24 months with monthly data. I have created a month/year variable called ymdate.

This has been my code so far:

bys perm : asreg Excess_USD_w MKT SMB HML, wind(ymdate 24)
gen residuals = Excess_USD_w - _b_cons - _b_MKT*MKT - _b_SMB*SMB - _b_HML*HML
bys perm: egen IVOL=sd (residuals)

In the end, I want to look at each year. However, my min and max values are the same for multiple years, not just for two years as specified by my rolling window. In addition, my mean value just keep on increasing.

Would you know what I have done wrong? Thank you very much in advance. I sincerely appreciate your help.

Best regards,
Kate

[Julien Maas]

Dear Professor Shah,

Somehow, asreg,fmb gives me the same output regardless of what window I am using is this normal?

Also I am getting high R2's while my coefficients from the in sample tests are mostly insignficant. Any suggestions?

Kind regards, Julien.

[Attaullah Shah]

Julien Maas
The window() option is not intended to be used with the fmb option. The window() option is designed for rolling regressions and does not work with the Fama and MacBeth regressions in asreg.

[Julien Maas]

Dear professor Shah,

Thank you for your swift response! That makes a lot of sense. So if I run;

asreg yearly_return pastreturn MVlog logBM, newey(16) fmb window(date -120 -1)

Then I will get slopes of the entire sample?

Also I was wondering what is the difference between using xtreg then in that case and asreg, as I get some very different results.

Furthermore as I am trying to perform in and out of sample tests similar to Lewellen's 2015 paper on a cross section of stock returns,

I was wondering,
If I first run (without the fmb);

asreg yearly_return pastreturn MVlog logBM, newey(16) window(date -120 -1)

And I get the beta's for each date, based on a ten year rolling return window,

What is then the best way to summarize the average of these ten year FM slopes for each signal. Displaying it's average slope, t-statistic, and R2. (Similar to table 8b of Lewellen's 2015 paper).

Thank you in advance for your time and consideration,

[Huong Pham]

Hi everyone, Im running a loop in Stata and found that asreg is really usefull. However, when I compare the two results, it is different, here are my code:
Using loop (traditional way):
forvalues i=1(1) 3 {
reg return market_ret if id==`i' & estimation_window==1
predict p if id==`i'
replace normal_return= p if id ==`i' & event_window==1
drop p
}

Using asreg:
bys id: asreg return market_ret, w(datenum 100) fit se

Do you know why the results are different?

[Huong Pham]

Option noconstant was added to asreg version 3.1.1 (26 July, 2018). However, that option works only with rolling window regressions and by-group regressions. I shall consider adding noconstant for option fmb in the next update.

Dear Prof. Shah,
Im running a loop in Stata and found that asreg is really usefull. However, when I compare the two results, it is different, here are my code:
Using loop (traditional way):
forvalues i=1(1) 3 {
reg return market_ret if id==`i' & estimation_window==1
predict p if id==`i'
replace normal_return= p if id ==`i' & event_window==1
drop p
}

Using asreg:
bys id: asreg return market_ret, w(datenum 100) fit se

Do you know why the results are different?
Thank you a lot for your time and your help!!

[Attaullah Shah]

Use the if statement with asreg as well. The window option in asreg implies a rolling window.

bys id: asreg return market_ret if estimation_window==1,  fit se

[Flynn Davies]

Dear Attaullah Shah

I was wondering if you had any experience running Fama Macbeth regressions with unbalanced panel data.

The data in question is unbalanced as I have firm fundamental information by year over the period 1987 - 2023. Naturally firms IPO at later dates or cease trading at various stages over this period.

I am running the following code;

logit DivPayer idiosyncratic_volatility M2B Asset_Growth Profitability SIZE

estimates store logit_model
estimates table logit_model, eform
xtset gvkey stata_date

newey DivPayer Legislation_Dummy idiosyncratic_volatility M2B Asset_Growth sqrt_profitability SIZE, lag(2)

However I get a

"year is not regularly spaced
r(198);"

error when attempting to run the code. Any idea how to resolve this issue?

Cheers,
Flynn