Ponente
Carlos Narciso Bouza Herrera
(Universidad de La Habana)
Descripción
The estimation of Differences-in-Differences (DiD) using a simple random sampling with replacement (SRSWR) sample is developed. The cases in which s is selected and it is s1 and s0 are determined deterministically or randomly are considered, as well as the case in which non-responses are present. Different alternative models are developed. Their behaviour in a real-life problem is discussed.
Key Words: Difference-in-Difference estimation, simple random sampling, expected error, non-responses, subsampling rules.
Autor primario
Carlos Narciso Bouza Herrera
(Universidad de La Habana)
