{"repo":"CFA-Institute-RPC/Explainable-AI-In-Finance","free":true,"listed":false,"github":"https://github.com/CFA-Institute-RPC/Explainable-AI-In-Finance","clone":"git clone https://github.com/CFA-Institute-RPC/Explainable-AI-In-Finance.git","description":"This repository contains accompanying code for the CFA Institute's Research and Policy Center 'Explainable AI in Finance: Addressing the Needs of Diverse Stakeholders' report.","language":"Jupyter Notebook","stars":18,"topics":["data-science","explainable-ai","quantitative-finance"],"license":"MIT","category":"trading","readme_excerpt":"🧠 Explainable AI in Finance: Fundamental Factor Modeling This repository explores how Explainable Artificial Intelligence (XAI) methods can be applied to machine learning models used for a case study in fundamental factor investing. It accompanies the CFA Institute's report on 'Explainable AI in Finance: Addressing the Needs of Diverse Stakeholders' (2025), illustrating the practical use of post-hoc explainability tools in financial modeling. 📊 Project Overview We implement a supervised learning workflow to predict monthly excess returns of stocks based on fundamental factor exposures using the XGBRegressor model from XGBoost. Our dataset contains normalized exposures to six BARRA-style factors across 218 stocks over a 100-month period starting in February 2008. We thank Bloomberg LP for use of the price and factor loadings data in this case study. 🧪 Workflow 1. Data Preprocessing The data provided has already been re-scaled. 2. Model Training - Split data into training and test sets - Use cross-validation to optimize hyperparameters - Train final model on full training set 3. Model Evaluation Assess predictive performance using root mean squared error (RMSE). 4. Explainability Methods We apply multiple XAI methods to understand model behavior: - Feature Importance : Gain-based feature ranking from XGBoost - Partial Dependence Plots (PDP) : Global sensitivity to features - Individual Conditional Expectation (ICE) : Instance-level visual insights - SHAP (SHapley Additive Ex","default_branch":null,"files":null,"tree":[],"storefront":"/r/CFA-Institute-RPC","claimed":false,"request_supported":{"post":"https://gitbuyer.com/r/CFA-Institute-RPC/Explainable-AI-In-Finance/request-supported","requests":0},"note":"indexed from public GitHub; nothing is for sale on this page. Clone it from GitHub. Paid listings live at /search."}