st: coldiag2 to check for collinearity.

[natasha agarwal <agarwana2@googlemail.com>]

Dear Everyone,

I was estimating an augmented production function on an unbalanced
panel data using a within estimator.

I issue the following commands

xtreg lnrval lnk lnw lnvfdi y14-y18, fe vce(robust)

Table 1. Intra-and/or-Inter-Industry  FDI-induced spillovers within a Province
Dependent Variable: Log of Value Added of Domestic Firms
	                  (1)	          (2)
Log of Capital	0.222***	0.204***
	              (0.011)	(0.011)
Log of Wages	0.561***	0.568***
	              (0.017)	(0.017)
Log of FDI	     0.294***	0.022
	              (0.013)	(0.020)
year = 2001	   -	      -0.326***
	                    -	       (0.017)
year = 2002	     -	    -0.231***
	                     -	     (0.013)
year = 2003	     -	-0.146***
	                    -	  (0.010)
year = 2004	   -	  -0.129***
	                -	(0.008)  		
		
Joint Significance of Time Dummies (F-test)	113.82***	


I understand that the highly significant coefficient on the time
dummies in the regression is because they are highly collinear with
lnvfdi as the results show above.

So I test for the same with the following command:

coldiag2 lnk lnw lnvfdi y14-y18

Condition number using scaled variables =   4028100.06

Condition Indexes and Variance-Decomposition Proportions

    condition
        index    _cons      lnk      lnw   lnvfdi      y14      y15
  y16      y17      y18
1        1.00      .        .        .        .        .        .
  .        .        .
2        2.21      .        .        .        .        .        .
  .        .        .
3        2.21      .        .        .        .        .        .
  .        .        .
4        2.21      .        .        .        .        .        .
  .        .        .
5        2.21      .        .        .        .        .        .
  .        .        .
6        7.83      .       0.49      .        .        .        .
  .        .        .
7       18.09      .       0.31      .       0.66      .        .
  .        .        .
8       24.99      .        .       0.74     0.30      .        .
  .        .        .
9  4028100.06     1.00      .        .        .       1.00     1.00
 1.00     1.00     1.00

Variance-Decomposition Proportions less than .3 have been printed as "."

Looking at the time dummies equal to 1 in the variance decomposition
proportions, can anyone help me explain does this mean that they are
highly collinear in the model and the same should be excluded from the
same?

Thanks
Natasha
*
*   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/