Data imputation in two-level factorial experiments
Missing data; Incomplete experiment; Fractional factorial; Estimation
method.
Full and fractional factorial designs with two levels are widely used in various fields
of knowledge, especially in industry. To analyze such experiments is necessary that all
planned treatment combinations are performed and the answers are obtained. However,
in practice, many experiments fail to be completed due to logistical problems, time or
budget constraints. These experiments are called incomplete. To properly analyze
such experiments, different methods are proposed in the literature. This study aims to
present, compare and make critical reflections about methods for estimating missing
data in factorial experiments with two levels.