For network DEA, no studies mentioned this relationship. So we can go on with the previous studies results till any update is found.

Dear Boru,

Banker et al. (1989), suggest a rough rule of thumb. Let p be the number of inputs and q be the number of outputs used in the analysis, then the sample size (n) should satisfy n >= max {p × q, 3(p + q)}.

Thank you

For network DEA, no studies mentioned this relationship. So we can go on with the previous studies results till any update is found.

Take a look at the following articles;

1. https://www.academia.edu/15284036/TESTS_OF_EFFICIENCY_IN_DATA_ENVELOPMENT_ANALYSIS

2. https://www.academia.edu/30342682/Sensitivity_analysis_of_inefficient_units_in_data_envelopment_analysis

3. https://www.academia.edu/18709973/Application_of_data_envelopment_analysis_.....................

4. Article Data Envelopment Analysis in Stata

5. https://www.academia.edu/1391030/Data_Envelopment_Analysis_in_Stata

Some literature review can help in this regard.