Proofs and additional experiments on Second order techniques for learning time-series with structural breaks

TL;DR

This paper offers detailed proofs of lemmas related to second order learning techniques for time-series with structural breaks and presents experimental results validating these methods.

Contribution

It provides complete proofs of key lemmas and experimental validation for second order techniques in time-series with structural breaks.

Findings

Proofs of lemmas about regularized loss functions

Experimental results supporting technique validity

Validation of second order methods for structural breaks

Abstract

We provide complete proofs of the lemmas about the properties of the regularized loss function that is used in the second order techniques for learning time-series with structural breaks in Osogami (2021). In addition, we show experimental results that support the validity of the techniques.