A0558
Title: Conditional latent factor models via model-based neural networks
Authors: Hao Ma - Queen Mary University of London (United Kingdom) [presenting]
Abstract: A hybrid methodology incorporating an econometric identification strategy is developed into artificial neural networks when studying conditional latent factor models. The time-varying betas are assumed to be unknown functions of numerous firm characteristics, and the statistical factors are population cross-sectional OLS estimators for given beta values. Hence, identifying betas and factors boils down to identifying only the function of betas, equivalent to solving a constrained optimization problem. For estimation, neural networks are constructed and customized to solve the constrained optimization problem, which gives a feasible non-parametric estimator for the function of betas. Empirically, the analysis is conducted on a large unbalanced panel of monthly data on US individual stocks with around $30,000$ firms, $516$ months, and $94$ characteristics. It is found that 1) the hybrid method outperforms the benchmark econometric method and the neural networks method in terms of explaining out-of-sample return variation, 2) betas are highly non-linear in firm characteristics, 3) two conditional factors explain over $95\%$ variation of the factor space, and 4) hybrid methods with literature-based characteristics (e.g., book-to-market ratio) outperform ones with COMPUSTAT raw features (e.g., book value and market value), emphasizing the value of academic knowledge from an angle of man vs. machine.
