By: Eric Siegel, Program Co-Chair, Predictive Analytics World for Financial
In anticipation of his upcoming conference presentation at Predictive Analytics World for Financial Las Vegas, May 
31-June 4, 2020, we asked Chakri Cherukuri, Senior Quantitative Researcher at Bloomberg, a few questions about their deployment of predictive analytics. Catch a glimpse of his presentation, Applied Machine Learning in Finance – Portfolio Management and Automated Trading, and see what’s in store at the PAW Financial conference in Las Vegas.
Q: In your work with predictive analytics, what behavior or outcome do your models predict?
A: This depends on the type of problem we are trying to solve. For example, in an unsupervised problem setting we are trying to understand the structure of high dimensional data through a low dimensional representation (for example, modeling yield curve dynamics using PCA or auto-encoders). In a supervised problem setting, we are trying to predict the label (either categorical or numerical) from training data. Training data can be time series based (stock/factor returns etc.) or text (news stories, financial documents etc.). We are also looking at pricing derivatives using machine learning methods where the features are market data plus deal parameters and the output is the price of the derivative.
Q: How does predictive analytics deliver value at your organization — what is one specific way in which it actively drives decisions or operations?
A: :There’s a big focus on delivering predictive analytics based on alternative datasets. One example is social sentiment extracted from news stories and tweets. We train machine learning models on this text data and provide company level sentiment scores to our clients who can then use these scores to build quantitative trading strategies.
Q: What surprising discovery or insight have you unearthed in your data?
A: Constructing labeled datasets (especially when dealing with alternative data) can be a challenge. These labels have to be manually annotated by people with domain knowledge and cannot be automated completely. It’s important to quality check the annotated data before training the machine learning models.
Q: Sneak preview: Please tell us a take-away that you will provide during your talk at Predictive Analytics World.
A: I’ll discuss machine learning based factor trading strategies. I’ll also compare these novel approaches to classical portfolio construction approaches.
—————————–
Don’t miss Chakri’s presentation, Applied Machine Learning in Finance – Portfolio Management and Automated Trading, at PAW Financial on Tuesday, June 2, 2020 from 4:20 to 4:40 PM. Click here to register for attendance.
By: Eric Siegel, Program Co-Chair, Predictive Analytics World for Financial
This content is restricted to site members. If you are an existing user, please log in on the right (desktop) or below (mobile). If not, register today and gain free access to original content and industry news. See the details here.
 |
 |
 |
