Binary Response Forecasting under a Factor-Augmented Framework

2025-07-22

AI Summary

This paper addresses the low predictive accuracy of binary response models (e.g., recession onset) under high-dimensional covariates. We propose a factor-augmented probit model that integrates latent factor analysis for dimension reduction and extraction of common variation, systematically embedding high-dimensional predictors into a binary regression framework. Theoretically, we establish asymptotic normality and consistency of the estimators. Methodologically, estimation proceeds via maximum likelihood, with Monte Carlo simulations confirming finite-sample robustness. Empirically, applied to U.S. recession forecasting, the model significantly outperforms the standard probit model both in-sample and out-of-sample—particularly improving the timing accuracy of recession identification. Our key contribution is the first theoretically rigorous, high-dimensionally adaptive predictive framework for binary outcomes that jointly leverages factor structure and probit modeling.

Technology Category

• Reasoning under Uncertainty: Relational Probabilistic Models
• Machine Learning: Probabilistic Circuits and Graphical Models
• Intelligent Robots: State Estimation

Application Category

• User Modeling, Personalization and Recommendation: Psychology-informed user models and recommender systems
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

In this paper, we propose a novel factor-augmented forecasting regression model with a binary response variable. We develop a maximum likelihood estimation method for the regression parameters and establish the asymptotic properties of the resulting estimators. Monte Carlo simulation results show that the proposed estimation method performs very well in finite samples. Finally, we demonstrate the usefulness of the proposed model through an application to U.S. recession forecasting. The proposed model consistently outperforms conventional Probit regression across both in-sample and out-of-sample exercises, by effectively utilizing high-dimensional information through latent factors.

Problem

• Forecasting binary response using factor-augmented regression
• Developing maximum likelihood estimation for regression parameters
• Improving recession prediction accuracy over Probit models

Innovation

• Factor-augmented binary response forecasting model
• Maximum likelihood estimation for regression parameters
• Utilizing high-dimensional information via latent factors