XGBoost

XGBoost (Extreme Gradient Boosting) is a highly popular and flexible gradient boosting library that implements machine learning algorithms under the gradient boosting framework (Chen and Guestrin, 2016). It is widely used for structured/tabular data due to its ability to handle complex nonlinear relationships, missing values, and its support for regularization to reduce overfitting.

Key Features

Highly flexible and configurable.
Supports L1 (Lasso) and L2 (Ridge) regularization.
Handles missing data internally.
Efficient and optimized for both memory and computation.
Strong community support and extensive tutorials.

Official Documentation

XGBoost Documentation

Notes

In this study, XGBoost was chosen for its proven reliability and flexibility, making it a strong baseline for gradient boosting on our dataset (~1,300 observations).