xtoverid, overidentification statistic not reported

[Liang Xu]

Dear Stata communities,

I am using the state command xtoverid to test the validity of my instrument. The regression commend is:
xtivreg y (x=z1 z2) x1,re vce(cluster firm)

The estimation is fine and the summary of the output is shown below:

See the attached pictures.

However, when I try to ran xtoverid, it gives a warning message, and the test statistics is missing as shown below:

xtoverid, robust

Warning: estimated covariance matrix of moment conditions not of full rank.
overidentification statistic not reported, and standard errors and
model tests should be interpreted with caution.
Possible causes: number of clusters insufficient to calculate robust covariance matrix
singleton dummy variable (dummy with one 1 and N-1 0s or vice versa)
partial option may address problem.

Test of overidentifying restrictions:
Cross-section time-series model: xtivreg g2sls robust cluster(firm)
Sargan-Hansen statistic . Chi-sq(.) P-value = .


I also tried xtoverid, cluster(firm) and xtoverid. The same warning message is shown and the test statistics is missing.

Do anyone know how to solve this problem? Thank you very much!!!

[Liang Xu]

The estimation results:

[Carlo Lazzaro]

Liang:
singleton dummy variable(s) might be the issue.
That said, I would consider a more parsimonios model.
As per FAQ, please do not post screenshots, because they are hard to read and impossible to elaborate on. Thanks.