simNA {meanscore} | R Documentation |
For this dataset, we generated 1000 observations of the predictor variable (X) from the standard normal distribution. The response variable(Y) was then generated as a Bernoulli random variable with p=exp(x)/(1+exp(x))
A dichotomous surrogate variable for X, called Z, was generated as follows:
Z=1, X >0
0, otherwise
We randomly deleted 500 of the X values (replacing them with NA), and stored the data in the matrix simNA, described below.
data(simNA)
There are 3 columns in the dataset.
Column 1 is the response variable (Y),
Column 2 is the surrogate variable (Z)
Column 3 is the predictor variable (X)