Reproduce “On the Identifiability of Regime-Switching Models with Multi-Lag Dependencies” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Carles Balsells-Rodas, Toshiko Matsui, Pedro A.M. Mediano, Yixin Wang, Yingzhen Li
The authors establish identifiability for multi-lag switching models by using overlapping sequence histories to distinguish regime trajectories and transition functions. For continuous latent-state models, further assumptions identify latent structure and regime-dependent lagged graphs. In synthetic tests, iSDS achieves higher regime-recovery F1 than compared methods across six settings, though performance weakens as regime count increases; these guarantees depend on stated assumptions.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.