get_type: Get the type of variables.

In hmi: Hierarchical Multiple Imputation

Description

Function checks of which type a variable is. The types are listed below (together with a rough summary of the classification mechanism).

• continuous (numeric values, or integers with more than 20 different values),

• semicontinuous (numeric values with more than 10% of them share the same value),

• rounded continuous (if more than 50% of the observations of a continuous variable are divisible by some rounding degrees)

• count data (integers).

• an intercept (the same value for all observations),

• binary (two different values - like 0s and 1s or "m" and "f"),

• categorical (the variable is a factor or has more than 3 different values)

• ordered categorical (the categorical variable is ordered.)

Usage

get_type(variable, spike = NULL, rounding_degrees = c(1, 10, 100, 1000))

Arguments

variable	A variable (vector) from your data set.
spike	A numeric value, denoting the presumed spike of semi-continuous variables.
rounding_degrees	A numeric vector with the presumed rounding degrees. If the rounding_degrees are set to be NULL, the proposed rounding degrees from the function suggest_rounding_degrees are used.

Value

A character denoting the type of variable.

Examples

get_type(iris$Sepal.Length); get_type(iris$Species)

hmi documentation built on Oct. 23, 2020, 7:31 p.m.