Specification LASSO and a Flexible Characteristics-Based Asset Pricing Model

Paper: arXiv 2611.03391

Authors: Chaohua Dong, Shuyi Ge, Shaoran Li, Wen Su

Abstract

This paper studies a partially linear semiparametric additive model that allows for interactions among variables in high-dimensional settings. We propose the Specification LASSO (S-LASSO), a two-step method that integrates LASSO and Adaptive Group LASSO to simultaneously perform variable selection and model specification. Theoretically, we establish that S-LASSO possesses oracle properties in terms of selection consistency and asymptotic normality. Monte Carlo simulations confirm that S-LASSO outperforms existing alternatives across a variety of scenarios. Empirically, we apply S-LASSO to a characteristics-based asset pricing model, exploring the linear and nonlinear effects of asset-specific characteristics on cross-sectional returns, as well as their interactions with firm size.

Complexity vs Empirical Score

• Math Complexity: 8.5/10
• Empirical Rigor: 6.0/10 (capped from 7.5: transaction costs not modelled; see penalty flags)
• Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: The paper introduces a sophisticated statistical method (S-LASSO) with strong theoretical backing, including oracle properties and asymptotic normality. It applies this method to a relevant problem in quantitative finance, demonstrating empirical superiority through simulations and a real-world asset pricing application. The combination of advanced mathematical derivations and rigorous empirical validation places it firmly in the ‘Holy Grail’ quadrant.

Research Flowchart

  flowchart TD
    A[Research Goal: Develop a Flexible Asset Pricing Model & Select Key Characteristics] --> B{Proposed Methodology: Specification LASSO (S-LASSO)};
    B --> C[Methodology Details: Two-step - LASSO for variable selection, Adaptive Group LASSO for model specification];
    C --> D[Theoretical Validation: Oracle Properties (Selection Consistency & Asymptotic Normality)];
    C --> E[Empirical Application: Characteristics-Based Asset Pricing Model];
    E --> F[Data Input: Asset-specific characteristics, Cross-sectional returns, Firm size];
    D & F --> G{Computational Process: Monte Carlo Simulations & Empirical Estimation};
    G --> H[Key Findings: S-LASSO outperforms alternatives; Identifies linear, nonlinear, and interaction effects of characteristics on returns];