"Economic Models or Machine Learning Techniques? Evidence from Asset Pricing Models" [SSRN]
Abstract: In this paper I compare popular machine learning techniques with traditional factor models using the level of mispricing as the primary evaluation metric. In making this comparison I highlight where machine learning models cannot be implemented in the traditional asset pricing framework. I also compare models based on forecast accuracy as measured by the out-of-sample r-squared . Traditional factor models achieve smaller average levels of mispricing and more accurate forecasts of asset returns. The benefits of deep learning recently documented in the literature appear the result of the empirical implementation of specific and not a universal truth.
"A Bankruptcy Risk Factor" [Submitted] [SSRN]
Abstract: This paper introduces a factor based on an estimated probability of bankruptcy - a measure of the risk a typical investor will lose their investment, or the cost of insuring that investment. Using an underlying model of firm bankruptcy built as a sequence of two random forests, I demonstrate my bankruptcy risk factor has predictive power in equity, bond, option and credit default swap markets earning statistically significant monthly returns of 0.23%, 0.15%, 1.97% and 1.04%, respectively, in all four markets. In markets with existing common factors I find statistically significant alpha with respect to these factors.
"Municipal Debt and the Equity Cross-Section of States" [Submitted] [SSRN]
Abstract: This paper develops a cross-sectional neural network for municipal bond yields, filling a need to extrapolate yields on these illiquid assets. Using this neural network, counterfactual municipal bond yields are generated for all states such that the only difference in the bond issues is the year and state where the issue took place. Hypothesizing that states with higher municipal bond yields present firms with riskier business environments, I demonstrate a mean-variance investor would be indifferent between a strategy of going long firms incorporated in high yield states and holding the S&P 500 index, but that absolute returns to this strategy dwarf buy-and-hold returns by more than 67%.
"Bankruptcy Risk and the Cross-Section of REITs" [Submitted] [SSRN]
Abstract: This paper investigates the equity cross-section of real estate investment trusts (REITs) both when REITs are added as a standalone portfolio to the cross-section of industries and when individual REITs are studied in isolation. A nine-factor asset pricing model which critically relies on the bankruptcy risk factor of Neumann (2021b) produces REIT portfolios which outperform the REIT market in terms of Sharpe ratio and the S&P 500 index in terms of absolute returns. The decrease in adjusted r-squared of an asset pricing model when REITs are included as a standalone portfolio is presented as an alternative quantification of the low correlation between REITs and other equity assets.
Select Non-Peer Reviewed Research
"Yield Predictions for New Jersey's Fiscal 2021 Bond Issues" - with Bruce Mizrach [ResearchGate]
"The Demise of the Mega Malls: A View from the Bond Market" - with Max Miller and Bruce Mizrach [ResearchGate]
"Revenue Predictions for the New Jersey Shore" - with Max Miller and Bruce Mizrach [ResearchGate]