Reference sources

Reference material used for orientation; read primary and academic sources first when claims conflict.

1. [S1] Reference

   Friedman (2001) -- Greedy function approximation: a gradient boosting machine. Introduced functional gradient descent, pseudo-residuals, and shrinkage.

   Reference

2. [S2] Reference

   Friedman (2002) -- Stochastic gradient boosting. Added row subsampling at each round as additional regularization.

   Reference

3. [S3] Reference

   Chen and Guestrin (2016) -- XGBoost: a scalable tree boosting system. Second-order Taylor expansion, regularized objective, sparsity-aware splits, system-level optimizations.

   Reference

4. [S4] Reference

   Ke, Meng, Finley, Wang, Chen, Ma, Ye, Liu (2017) -- LightGBM: a highly efficient gradient boosting decision tree. Gradient-based one-side sampling (GOSS) and exclusive feature bundling (EFB).

   Reference

5. [S5] Reference

   Prokhorenkova, Gusev, Vorobev, Dorogush, Gulin (2018) -- CatBoost: unbiased boosting with categorical features. Ordered boosting to prevent target leakage.

   Reference

6. [S6] Reference

   Gu, Kelly, Xiu (2020) -- Empirical asset pricing via machine learning. Large-scale comparison finding gradient-boosted trees competitive with neural networks.

   Reference

7. [S7] Reference

   Medeiros, Vasconcelos, Veiga, Zilberman (2021) -- Forecasting inflation in a data-rich environment. Boosted trees outperforming linear penalized models for Brazilian macro.

   Reference