Random Forests for Regression with Missing Entries

These are specific codes used in the article:

• On the Consistency of a Random Forest Algorithm in the Presence of Missing Entries, Irving Gómez-Méndez and Emilien Joly, 2021 (to be published).
• Regression with Missing Data, a Comparison Study of Techniques Based on Random Forests, Irving Gómez-Méndez and Emilien Joly, 2021 (to be published).

Examples on the use of these codes as well as updated versions can be found in https://github.com/IrvingGomez/Random_forests_with_missing_values

In Source it can be found "random_forests_with_missing_values.R" which contains the needed functions to construct the random forests with missing values using the approach proposed in "On the consistency of a random forest algorithm in the presence of missing entries".

The file Create_datasets contains the codes to create the training datasets and the testing datasets. While the file Create_RF_and_Predict_RF contains the codes to create the Random Forests using our approach and to predict new observations with missing values accordingly to the procedure describe in "Regression with Missing Data, a Comparison Study of Techniques Based on Random Forests"

Finally, the file Calculate_MSE contains the codes to calculate the MSE.

License: All the code is under the GNU GPL v3 license or any posterior version.

©️ (21-05-2021) Irving Gómez Méndez.