Novel MNAR Multivariate mechanism

A novel example of generate artificial missing data with mdatagen library with the Breast Cancer Wiscosin dataset from scikit-learn. The features will receive the missing values under Missing Not at Random (MNAR) mechanism. The simulated missing rate is 20%. The method to choose missing values is Missingness Based on Own and Unobserved Values (MBOUV).

In [1]:

# Import the libraries
import pandas as pd
from sklearn.datasets import load_breast_cancer

from mdatagen.multivariate.mMNAR import mMNAR

# Load the data
wiscosin = load_breast_cancer()
wiscosin_df = pd.DataFrame(data=wiscosin.data, columns=wiscosin.feature_names)

X = wiscosin_df.copy()   # Features
y = wiscosin.target    # Label values

# Create a instance with missing rate equal to 20% in dataset under MNAR mechanism
generator = mMNAR(X=X, y=y)

# Generate the missing data under MNAR mechanism
generate_data = generator.MBOUV(missing_rate=20, depend_on_external=X.columns)
print(generate_data.isna().sum())

# Import the libraries import pandas as pd from sklearn.datasets import load_breast_cancer from mdatagen.multivariate.mMNAR import mMNAR # Load the data wiscosin = load_breast_cancer() wiscosin_df = pd.DataFrame(data=wiscosin.data, columns=wiscosin.feature_names) X = wiscosin_df.copy() # Features y = wiscosin.target # Label values # Create a instance with missing rate equal to 20% in dataset under MNAR mechanism generator = mMNAR(X=X, y=y) # Generate the missing data under MNAR mechanism generate_data = generator.MBOUV(missing_rate=20, depend_on_external=X.columns) print(generate_data.isna().sum())

mean radius                166
mean texture                61
mean perimeter              35
mean area                  148
mean smoothness            156
mean compactness           161
mean concavity             112
mean concave points        171
mean symmetry               93
mean fractal dimension     110
radius error                74
texture error              160
perimeter error             82
area error                  63
smoothness error           160
compactness error          152
concavity error             99
concave points error       143
symmetry error             171
fractal dimension error      1
worst radius               170
worst texture              159
worst perimeter              4
worst area                 170
worst smoothness            51
worst compactness          148
worst concavity             42
worst concave points       126
worst symmetry              78
worst fractal dimension    148
target                       0
dtype: int64

In [ ]: