st: Re: Normality and Granger Causality in Panel Data

["D. Demetriou" <dd358@cam.ac.uk>]

Kit and Nick,

Thank you for all your comments and suggestion on how to deal with 
normality and granger causality in Panel Data stractures.

Chrostopher in his last email suggested that because my panel is short(T=10 
and N=9) I should't use a cluster-robust estimator. However, STATA does 
implemented onw automatically as you can see below the heading just above 
the estimation table. is there any way to disable this??


. xi: xtreg publicprov1 hownership incvars popabove60 ILO regsegr 
primsector i.wave
> , fe vce(robust)
note: w10 omitted because of collinearity

Fixed-effects (within) regression Number of obs = 80 Group variable: region 
Number of groups = 9

R-sq: within = 0.9154 Obs per group: min = 8
      between = 0.1848 avg = 8.9
      overall = 0.2093 max = 9

                                               F(8,8) = . corr(u_i, Xb) = 
-0.9213 Prob > F = .

                                (Std. Err. adjusted for 9 clusters in 
region) 
------------------------------------------------------------------------------
            |               Robust
publicprov1 | Coef. Std. Err. t P>|t| [95% Conf. Interval] 
-------------+----------------------------------------------------------------
 hownership | -.1460149 .041091 -3.55 0.007 -.2407709 -.0512588
    incvars | -.0324381 .025144 -1.29 0.233 -.0904203 .0255441
 popabove60 | .0431907 .0277384 1.56 0.158 -.0207741 .1071556
        ILO | .00225 .026478 0.08 0.934 -.0588084 .0633083
    regsegr | .0564663 .0228579 2.47 0.039 .0037559 .1091768
 primsector | -.0458808 .0555693 -0.83 0.433 -.1740238 .0822621


Yours,

DD
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/