-
Notifications
You must be signed in to change notification settings - Fork 0
FactorFactory: Financial Factor Replication
The FactorFactory (https://github.com/cdldl/FactorFactory) project draws inspiration from the seminal works on factor replication, notably those by Bryan Kelly in his Replication Crisis repository (https://github.com/bkelly-lab/ReplicationCrisis) and the enhancements introduced by Chen and Zimmermann's library (https://github.com/OpenSourceAP/CrossSection). These foundational works underscore the intricacies and challenges inherent in faithfully replicating financial factors. FactorFactory aims to build on these insights by offering a robust suite of factors for multiple non-academic data providers designed to facilitate the accurate replication, comprehensive analysis, and thorough validation of financial factors within the realm of quantitative finance. By simplifying these processes, FactorFactory endeavors to advance the field of financial research, making it more accessible for researchers, analysts, and practitioners to rigorously test theories, explore novel factor models.
While several R packages offer tools for financial analysis and factor modeling, they often cater to a broader array of financial analysis. FactorFactory seeks to bridge this gap by providing a focused array of functionalities specifically engineered for the nuanced process of financial factor replication.
The 12-week coding period will prioritize development in the following areas:
- Factor Construction: Development of functions to build widely recognized financial factors (e.g., Fama-French three-factor model) from scratch.
- Data Management: Creation of utilities for efficient data preprocessing, normalization, and alignment, particularly for financial time series data.
- Statistical Analysis: Implementation of robust statistical frameworks for the empirical testing of factor models, including backtesting methodologies (such as with the R package FactorAnalytics and ExpectedReturns), significance evaluations, and comparative performance analysis.
- Documentation for Chen and Zimmermann's naming convention
- Documentation for multiple data provider (i.e. eodhd.com, Bloomberg) field names to Bryan Kelly's naming convention
- Vignettes with test cases with FactorAnalytics R library
- Vignettes with test cases with ExpectedReturns R library
- Comprehensive unit testing for every major function and to test factors to ensure code quality and reliability.
- Validate tests to create an R package on CRAN
By introducing the FactorFactory package to the R ecosystem, we aim to significantly enhance the toolkit available for financial factor research. This project will democratize access to sophisticated factor replication methodologies, empowering a broader segment of the academic and professional finance community to conduct rigorous, reproducible research. FactorFactory is poised to set new standards for transparency and methodological rigor in financial research, fostering innovation and encouraging a deeper, more critical engagement with financial models and theories.
- EVALUATING MENTOR: Cyril de Lavergne [email protected]
- Other Mentors Justin Shea [email protected] and Brian G. Peterson [email protected]
- Make a documentation of the features in Chen and Zimmermann’s (https://github.com/OpenSourceAP/CrossSection) not available in Bryan Kelly’s.
- Visit the code of the library (https://github.com/cdldl/FactorFactory) suggest enhancements and code some of the factors in Chen and Zimmermann’s library not in Bryan Kelly’s documentation
- Provide a solid framework to make sure your new factors are correct (unit test).
Contributors are encouraged to share their test solutions at [email protected]