Warning: one more more cross-sections have been dropped due

[tvonbrasch]

Hi,

I am estimating an equation using DOLS in a panel. the equation gives the warning:
Warning: one more more cross-sections have been dropped due to estimation errors

What does it mean? In particular, what does estimation errors mean? Also, what does the negative R-squared -114.590409 mean?
Thomas



Dependent Variable: LOG(TFPR)
Method: Panel Dynamic Least Squares (DOLS)
Date: 06/17/14 Time: 19:28
Sample (adjusted): 1976 2009
Periods included: 34
Cross-sections included: 12
Total panel (unbalanced) observations: 363
Panel method: Weighted estimation
Cointegrating equation deterministics: C
Fixed leads and lags specification (lead=1, lag=1)
Long-run variance weights (Prewhitening with lags = -1 from AIC
maxlags = -1, Bartlett kernel, Newey-West fixed bandwidth)
Warning: one more more cross-sections have been dropped due to
estimation errors

Variable Coefficient Std. Error t-Statistic Prob.

LOG(RT) 0.045338 0.007280 6.227374 0.0000
LOG(RT)*LOG(AT) 0.173740 0.021512 8.076408 0.0000
H_HS 2.994684 0.227814 13.14532 0.0000

R-squared -114.590409 Mean dependent var -0.343002
Adjusted R-squared -142.805391 S.D. dependent var 0.427470
S.E. of regression 5.126168 Sum squared resid 6674.510
Long-run variance 0.035784

[EViews Glenn]

The R2 measure is a bit strange in this case because of the weighting. I'd ignore it. As to the dropped, it's hard to say without seeing the data, but the idea is that each cross-section has to be detrended and then weighted. In the course of doing so there can be errors due to singularity or problems with computing the weights. In that case, the cross-section is dropped from estimation and a warning is issued.