Decomposing Rank Reversals under Temporal Nonstationarity: Alignment, Aggressiveness, and Instability
cs.LG, q-fin.ST
Submitted: 2026-11-03
Updated: 2026-11-03
License: http://creativecommons.org/licenses/by/4.0/
The gist: Under temporal nonstationarity, model performance rankings reverse across training horizons due to heterogeneous sensitivities to memory.
Abstract
Under temporal nonstationarity, model performance rankings reverse across training horizons due to heterogeneous sensitivities to memory. While existing literature addresses adaptive lookbacks, effective model complexity, and empirical forecast evaluation in isolation, no unified framework explains the structural mechanics driving these rank reversals. We bridge this gap by deriving a model-agnostic decomposition of out-of-sample R squared into target alignment, miscalibrated aggressiveness, and predictive instability. Within this framework, we mathematically characterize complexity phase boundaries, establish closed-form optimal-memory equations for ridge regression under local drift, and extend these guarantees to complex model architectures via surrogate bounds. We validate these theoretical dynamics by benchmarking 14 linear, tree-based, and neural architectures across large-scale financial panels. In doing so, our framework demonstrates that empirical model rank reversals are driven by predictable surges in miscalibrated aggressiveness, providing a rigorous mechanistic diagnosis of generalization under temporal shift.
Sources
- Adaptive Estimation and Learning under Temporal Distribution Shift
- The Nonstationarity-Complexity Tradeoff in Return Prediction
- Time-varying Forecast Combination for High-Dimensional Data
- TempusBench: An Evaluation Framework for Time-Series Forecasting
- A Look at the Evaluation Setup of the M5 Forecasting Competition
- A Stability Principle for Learning under Non-Stationarity
- How do standard time series forecasting benchmarks translate to finance? A broad study of risk adjusted performance
- Benchmarking Deep Time Series Models for Equity Portfolios
Related papers
- A Physics-Regulated Neural Framework for Learning 3D Grain Growth Dynamics
- FlexMoE: Learning One-for-All Multi-Budget Expert Pruning for Flexible MoE Deployment
- OptiMind: Improving Small Language Models for Optimization Modeling through Formulation Guidance and Multi-Turn Learning
- Staying in the Sweet Spot: Boost Reasoning Evolution via Responsive Rollout Difficulty Manipulation
- Examining the Impact of Data and Training Methods on Emergent and Subliminal Misalignment
- Rethinking Black-Box On-Policy Distillation: A Rubric-based Analysis