The Curious Quant EP12: Vinesh Jha: The craft of mining alternative data
Feb 5, 2020
Vinesh Jha, CEO of ExtractAlpha, brings his expertise in quantitative research and prop trading to discuss the world of alternative data in finance. He delves into the complexities of stock selection and the necessity of distinguishing signal from noise. The conversation explores the evolution of data analysis methods, including sentiment analysis and market behaviors. Vinesh emphasizes the integration of diverse data sources to enhance investment strategies, while contemplating the challenges of predicting financial returns and the future role of AI in finance.
AI Snips
Chapters
Books
Transcript
Episode notes
Career Path From Research To Startup
- Vinesh Jha described his career from sell-side equity research to founding Extract Alpha in Hong Kong.
- He highlighted experience at StarMine, Merrill Lynch and Morgan Stanley's Process Driven Trading group before launching his firm in 2013.
Opinions Are Diverse Data Signals
- Opinions are a broad alternative-data class that include sell-side estimates, crowdsourced forecasts, options and insider trades.
- Different opinion sources offer distinct signals and motivations which can be exploited separately.
Predict Fundamentals First
- Predict fundamentals (revenues, earnings surprises) with alternative data instead of directly predicting returns.
- Use those fundamental forecasts as more robust building blocks for price models rather than chasing noisy short-term returns.
