irrelevant: Class "irrelevant" and Inherited Classes

In mi: Missing Data Imputation and Model Checking

Class "irrelevant" and Inherited Classes

Description

The irrelevant class inherits from the missing_variable-class and is used to designate variables that are excluded from the models used to impute the missing values of “relevant” variables. For example, if a survey has an “id” variable that simply distinguishes observations, the user should designate it as irrelevant, although it will automatically be classified so if its name is either “id” or starts with punctuation (including underscores). The fixed class inherits from the irrelevant class and is used for variables that are constant (within a sample). A variable that is instantiated from the fixed class cannot have any missing values. The group class inherits from the fixed class and is used like a factor to spit samples in multilevel modeling; see multilevel_missing_data.frame-class. None of these classes have an additional slots. Aside from these facts, the rest of the documentation here is primarily directed toward developeRs.

Objects from the Classes

The missing_variable generic function can be used to instantiate an object that inherits from the irrelevant class by specifying type = "irrelevant", type = "fixed", or type = "group".

Author(s)

Ben Goodrich and Jonathan Kropko, for this version, based on earlier versions written by Yu-Sung Su, Masanao Yajima, Maria Grazia Pittau, Jennifer Hill, and Andrew Gelman.

See Also

missing_variable-class

Examples

# STEP 0: GET DATA
data(nlsyV, package = "mi")

# STEP 0.5 CREATE A missing_variable (you never need to actually do this)
first <- missing_variable(as.factor(nlsyV$first), type = "group")
show(first)

mi documentation built on June 7, 2022, 1:04 a.m.