Millions way to avoid overfitting when building a portfolio


Chia-Yi Yen /英語


portfolio, quantitative finance, overfitting


You never know how to build a successful portfolio that always gives you significant return, but you definitely should know what would make you build a model that is doomed to be a failure. Overfitting is the most common one. In the brief talk, I would like to share some experience how to avoid overfitting. At least in some ways. The examples would be shown in Python toolkit like ipython, pandas, numpy, matplotlib, and seaborn.


Profile picture
Chia-Yi Yen, you can call her "Yen" just like the Japanese dollar yen, is a financial engineer in a local hedge fund. She has been responsible for multi-factor model construction, data infrastructure, data Extract-Transform-Load solution, and other numeric analysis projects. She is active in many open-source communities like Python and R, and she is also the organizer of Taiwan R-Ladies community, which is a female meetup of everything about data in the world of R language.


Risksoft Technology Ltd., Quantitative Strategies Department


Financial Engineer