crsp

Calculates 103 firm characteristics from CRSP + Compustat directly in Python – no WRDS SAS cloud

Sample SAS programs that process WRDS data and facilitate econometric analysis

Data matching for corporate governance research

A python script to create a mapping table between I/B/E/S and Compustat

This GitHub repository shows data collection and analysis for “Regulatory Fragmentation” paper by Kalmenovitz, Lowry and Volkova, The Journal of Finance (Forthcoming)

Fuzzy match entity names (primarily persons and companies) across databases

US-equity quantitative research, backtest and paper-trading system (codename Plutus).

Pipeline dealing with WRDS (Wharton Research Data Services) datasets including crsp, master, etc, in order to build mega-database for scaling in Market Microstructure research

Classify CRSP-style delisting reasons from SEC EDGAR for quant pipelines

Replication code for "Industry factor structure: Beta estimation and cross-sectional returns" (Woo & Kim, 2026)

End-to-End Python implementation of Mo et al.'s (2025) ACT-Tensor methodology; a tensor completion framework for financial dataset imputation. Implements cluster-based CP decomposition, HOSVD factor extraction, temporal smoothing (CMA/EMA/Kalman), and downstream asset pricing evaluation. Transforms sparse data into dense machine readable data.

Stock Factor Data

Rolling-window XGBoost cross-sectional return prediction for US equities (1995-2024). Out-of-sample annualized Sharpe 1.03, monthly CAPM alpha +2.19% (t=6.08), market beta -0.43 over 300 months (2000-2024).

Point-in-time insider filing de-noising, ML scoring, and cross-sectional return tests.

按决策难度匹配 Agent 介入方式的智能数据准备系统 | Intelligent Data Preparation Agent

Idiosyncratic volatility and abnormal returns during VIX spike events empirical study using a survivorship-bias-free CRSP universe (2005–2024)

Carhart 1997 Momentum UMD factor construction using CRSP monthly data. Size-momentum portfolios, decile analysis, Netflix tracking.

ML pipeline for monthly U.S. equity return prediction using CRSP / Compustat / JKP factor characteristics. Implements OLS, Ridge, Lasso, XGBoost, and MLP models with rolling-window evaluation and IC analysis.

Jump detection via expanding-window Z-score on SPY and BRK.B (2020-2025) + logistic regression using Loughran-McDonald NLP sentiment from 48 SEC 8-K filings. McFadden R²=0.50. VCU FIRE 691.

Testing stock return predictability using lagged dividend-price ratio (Cochrane / Goyal-Welch) on CRSP annual and monthly data (1927-2024). 5-year R² of 8.41%. VCU FIRE 691.