This course is a dense presentation of machine learning (ML) tools used in financial risk management, portfolio management, and trading. Ten classes are offered: two on risk management, two on loan portfolio management, three on portfolio optimization, and three on high-frequency trading. The risk classes cover the risk measurement of financial assets using distribution fitting, copulas, PCA, and splines. The loan portfolio management classes cover risk estimation and backtesting using logistic regression, regularization, clustering methods, and the applied statistics concepts such as parameter and process risk. Kaggle competitions for loan portfolios which used tree-based algorithms for predictions are also reviewed. The classes on portfolio optimization introduce classic theories for asset return estimation and their extensions (multi-factor models) while using unsupervised & supervised ML methods to verify & derive new factors; modern portfolio theory using constrained optimization & robust methods; and Black-Litterman model portfolios where asset-specific, ML-derived models are integrated. The classes on trading introduce the limit order book and market microstructure and then move on to tour the winning strategies of to Kaggle competitions on trading. The feature engineering and code of the winning solutions are reviewed in depth.
Where does machine learning show up in finance? Does it enhance portfolio analytics, risk analytics, or trading? By knowing where machine learning currently fits into these fields, you can be ready to incorporate enhancements as they are discovered. Machine learning techniques as subtle as out-of-sample testing can enhance portfolio optimization. Logistic regression combined with clustering algorithms can be used to not only predict risk of default, but more importantly, backtest and manage a loan portfolio. Machine learning competitions on Kaggle display a multitude of machine learning algorithms used for winning trading strategies or for superior credit risk estimation.
The goal is to provide a bridge from knowledge of machine learning, programming, and statistics to a foundational understanding of how those resources are applied to finance. And ambitiously, the course also strives to summarize some current practices and empower students to take the next steps in their development of any of the three main topics: risk, portfolio management, and trading.
Who Is This Course For?
The course can be a great learning experience for traders, risk managers, portfolio managers, investors, and those looking to build skills to work in the finance industry. It is designed for those who already have a foundation in Machine Learning and want to see the tools and concepts applied to finance. The class will alternate between Excel, R, and Python. Basic knowledge of coding, machine learning, and finance are assumed.
Learn more and register: https://nycdatascience.com/...