Model Selection when multiple imputation is used to protect confidentiality in public use data.
Satkartar K. KinneyJerome P. ReiterJames O. BergerPublished in: J. Priv. Confidentiality (2011)
Keyphrases
- model selection
- data sets
- data analysis
- database
- cross validation
- input data
- multivariate regression
- hyperparameters
- regression model
- missing data
- training data
- parameter estimation
- prior knowledge
- hypothesis tests
- statistical inference
- databases
- probability distribution
- multiple imputation
- statistical agencies
- error estimation
- bayesian methods
- machine learning
- incomplete data
- closed form
- clustering algorithm
- sample size
- high dimensional data