The first 300 rows belong to the database A, while the next 400 rows belong to the database B.
common to both databases (same encodings).
Gender is the only complete covariate.
Yb2 are the target variables of A and B respectively, summarizing a same information encoded in two different scales.
that summarize a same information saved in two distinct encodings, that is why,
missing in the database B and
Yb2 is missing in the database A.
A data.frame made of 2 overlayed databases (A and B) with 700 observations on the following 8 variables.
the database identifier, a character with 2 possible classes:
the target variable of the database A, stored as factor and encoded in 3 ordered levels:
[60-80] (the values related to the database B are missing)
the target variable of the database B, stored as integer (an unknown scale from 1 to 5) in the database B (the values related to A are missing)
a factor with 2 levels (
Male) and no missing values
a covariate of 3 classes stored as a character with 2% of missing values:
a factor with 4 levels and 5% of missing values: from
Dos 1 to
a covariate of 2 classes stored as a character and 10% of missing values:
NO for non smoker,
a numeric corresponding to the age of participants in years. This variable counts 5% of missing values
The purpose of the functions contained in this package is to predict the missing information on
in database A and database B using the Optimal Transportation Theory.
Missing information has been simulated to some covariates following a simple MCAR process.