Data envelopment analysis with missing values: An interval DEA approach

Measuring and explaining technical efficiency of dairy farms: a case study of smallholder farms in East Africa

“…However, in certain contexts, excluding such DMUs can go against the spirit in which the DEA is conducted (Kuosmanen, 2002;Smirlis, Maragos & Despotis, 2006;Womer et al, 2000). For instance, in the context of dairy farms in East Africa, the researchers do not expect all farms to have consistently similar structure of inputs and outputs.…”

Evaluation of performance of European cities with the aim to promote quality of life improvements

“…A total of 50 cities were eliminated from the original database of the Urban Audit program. [20] to replace missing outputs by zero, or the approach of Smirlis et al [29] that suggest the replacement of missing values by approximations in the form of interval in which the unknown missing values are likely to belong. For the purpose of this work, we have chosen to replace the missing observations with the minimum value observed in the database for that variable.…”

Optimum estimation of missing values in randomized complete block design by genetic algorithm

“…Chen et al [11] proposed to use a dynamically updating approximation method to maintain Incomplete Ordered Decision Systems with two cases of missing data, i.e., ''lost value'' and ''do not care''. Smirlis et al [54] introduced an approach based on interval DEA that allows the evaluation of the units with missing values along with the other units with available crisp data. In their study, the missing values are replaced by intervals in which the unknown values are likely to belong.…”