[Mustapha Younis]

Thanks Carlo,

I understand from your answer that I should not use Hausman test, but I don't understand why and at the same time how should I use xtoverid?

[Carlo Lazzaro]

Mustapha:
because, if you decide to go -robust- or clustered standard errors, -hausman- does not support them.
Coversely, if you use default standard errors, -hausman- works fine.

[Mustapha Younis]

Thanks Carlo,

Then should I go

 xtreg GDPgrowthannual Manufacturingvalueaddedannua Foreigndirectinvestmentneti GrosssavingscurrentUS InflationGDPdeflatorannual Hightechnologyexportsofma Populationgrowthannual Schoolenrollmentsecondary OilExporting c.Manufacturingvalueaddedannua#OilExporting, re  

estimates store random  

xtreg GDPgrowthannual Manufacturingvalueaddedannua Foreigndirectinvestmentneti GrosssavingscurrentUS InflationGDPdeflatorannual Hightechnologyexportsofma Populationgrowthannual Schoolenrollmentsecondary OilExporting c.Manufacturingvalueaddedannua# OilExporting, fe  

estimates store fixed  

Xtoverid

Are these steps right?

[Mustapha Younis]

Dear Carlo,

Xtoverid is not working with me. I got an error message r(198).

error . . . . . . . . . . . . . . . . . . . . . . . .  Return code 198
        invalid syntax;
        option __________ incorrectly specified;
        option __________ not allowed;
        __________ invalid;
        range invalid;
        __________ invalid obs no;
        invalid filename;
        __________ invalid varname;
        __________ invalid name;
        multiple by's not allowed;
        __________ found where number expected;
        on or off required;
        All items in this list indicate invalid syntax.  These errors
        are often, but not always, due to typographical errors.  Stata
        attempts to provide you with as much information as it can.
        Review the syntax diagram for the designated command.
        In giving the message "invalid syntax", Stata is not helpful.
        Errors in specifying expressions often result in this message.

[Carlo Lazzaro]

Mustapha:
what if you type:

xi: xtreg GDPgrowthannual Manufacturingvalueaddedannua Foreigndirectinvestmentneti GrosssavingscurrentUS InflationGDPdeflatorannual Hightechnologyexportsofma Populationgrowthannual Schoolenrollmentsecondary OilExporting c.Manufacturingvalueaddedannua#OilExporting, re vce(cluster panelid)
xtoverid

[Mustapha Younis]

Dear Carlo,

I got the same error message.

I have another question, Is this interaction term right?

  
 c.Manufacturingvalueaddedannua#OilExporting

It is an interaction between an independent variable and a dummy variable. I just want to make sure that the command is right.

[Carlo Lazzaro]

Mustapha:
1) just to get a clearer insight in what you're doing: you started with a clustered standard errors (SEs), then in your last examples you seemingly switched to default SES. Please clarify.
That said:
a) if you go default: go -hausman-;
b) if you go clustered, go -xtoverid-: the error you experienced is probably due to an omitted variable, that yiou should omit manually (see https://www.stata.com/statalist/arch.../msg00885.html).
2) you interaction code should be:

c.Manufacturingvalueaddedannua##i.OilExporting

[Mustapha Younis]

Dear Carlo,

I don't know what's the meaning of going default or going clustered. But generally I want to go default to use the hausman test.

[Carlo Lazzaro]

Mustapha:
- default standard errors=the standar errors that -xtreg-provides when you do not specify any -vce()-options;
- clustered standard errors=-vce(cluster panelid)-.

[Mustapha Younis]

Dear Carlo,

I think I didn't specify vce, if I am not wrong, but I want to go with Hausman test.
Do you have any comment on the econometric model?

[Carlo Lazzaro]

Mustapha:
the advice about -cluster()- concerned the apparent heteroskedasticity in the epsilon residual distribution in the graph you originally posted.
If you've changed your model specification and fixed heteroskedasticity, obviously default standard errors are enough and -hausman- is the way to go.
Your econometric model is as good as it gives a fair and true view of the data generating process.

[Mustapha Younis]

Thank you Carlo.

Yes, I want to specify it without cluster and I want to use the hausman.
I just used the epsilon residual distribution in the graph because the hettest wasn't working with xtreg and I didn't know how to see for heteroskedasticity. If there's another way to see heteroskedasticity visually without going cluster I will be grateful if you guided me in that.

Regards.

[Carlo Lazzaro]

Mustapha:
there's no need to go -cluster()- to visually inspect epsilon distribution for potential heteroskedasticity:

 use "https://www.stata-press.com/data/r16/nlswork.dta"
. xtreg ln_wage age, re

Random-effects GLS regression                   Number of obs     =     28,510
Group variable: idcode                          Number of groups  =      4,710

R-sq:                                           Obs per group:
     within  = 0.1026                                         min =          1
     between = 0.0877                                         avg =        6.1
     overall = 0.0774                                         max =         15

                                                Wald chi2(1)      =    3140.35
corr(u_i, X)   = 0 (assumed)                    Prob > chi2       =     0.0000

------------------------------------------------------------------------------
     ln_wage |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
         age |   .0185667   .0003313    56.04   0.000     .0179174    .0192161
       _cons |   1.120439   .0112038   100.01   0.000      1.09848    1.142398
-------------+----------------------------------------------------------------
     sigma_u |  .36972456
     sigma_e |  .30349389
         rho |  .59743613   (fraction of variance due to u_i)
------------------------------------------------------------------------------

. predict fitted, xb
(24 missing values generated)

. predict residual, e
(24 missing values generated)

. scatter residual fitted

.

[Mustapha Younis]

Thanks a lot Carlo.

I learned from you a lot.

Best Regards.