Entropy-Based Static Imputation
Context
- Entropy-Based Static Imputation is based on Static Imputation.
- Whereas the standard, non-entropy-based approaches randomly choose the sequence in which missing values are imputed within a row of data, the entropy-based methods select the order based on the conditional uncertainty associated with the unobserved variable.
- More specifically, missing values are imputed first for those variables that meet the following conditions:
- Variables that have a fully-observed/imputed Markov Blanket.
- Variables that have the lowest conditional Entropy, given the observations and imputed values.
- The advantages of the entropy-based methods are:
- The speed improvement over their corresponding standard methods.
- Their improved ability to handle datasets with large proportions of missing values.
Usage
- Select
Menu > Learning > Missing Values Processing > Entropy-Based Static Imputation
.
For a thorough introduction to this topic, please see Chapter 9: Missing Values Processing in our e-book.