About the Event
In this talk, Rohit Allena, Ph.D., will discuss how to estimate standard errors or confidence of intervals of stock return forecasts that are based on various Machine Learning models, such as Neural Networks, Lasso, Elastic Net, and Ridge regression. Based on the precision of return forecasts, he will discuss how to construct novel “confidence-high-low" portfolios that outperform existing trading strategies. The confidence-high-low portfolios take long and short positions exclusively on subsets of stocks that have more precise return forecasts. He will also discuss why and how the confidence intervals around return forecasts relate to various macroeconomic variables.
About the Speaker
Allena joined the faculty at the C.T Bauer College of Business in fall 2021. He holds a Finance Ph.D. from Emory University, where his dissertation is comprised of three essays on estimation uncertainty. One of his essays won a Cubist Systemic Strategies award for outstanding research at the Western Finance Association annual meetings, held in July 2020. His research interests are in asset pricing and market-microstructure with an emphasis on the econometrics of Machine Learning inferences and Bayesian inferences.
