Deep LPPLS: Forecasting of temporal critical points in natural, engineering and financial systems

TL;DR

This paper introduces deep learning calibration techniques for the LPPLS model, significantly improving the accuracy and speed of predicting critical transition points in natural, engineering, and financial systems.

Contribution

It presents the Mono-LPPLS-NN and Poly-LPPLS-NN models, novel deep learning architectures that outperform existing methods in estimating LPPLS parameters and reduce computation time.

Findings

M-LNN outperforms state-of-the-art in parameter estimation accuracy.

P-LNN generalizes to unseen time-series and accelerates parameter estimation.

Models successfully predict financial bubbles and a rockslide transition.

Abstract

The Log-Periodic Power Law Singularity (LPPLS) model offers a general framework for capturing dynamics and predicting transition points in diverse natural and social systems. In this work, we present two calibration techniques for the LPPLS model using deep learning. First, we introduce the Mono-LPPLS-NN (M-LNN) model; for any given empirical time series, a unique M-LNN model is trained and shown to outperform state-of-the-art techniques in estimating the nonlinear parameters $(t_c, m, \omega)$ of the LPPLS model as evidenced by the comprehensive distribution of parameter errors. Second, we extend the M-LNN model to a more general model architecture, the Poly-LPPLS-NN (P-LNN), which is able to quickly estimate the nonlinear parameters of the LPPLS model for any given time-series of a fixed length, including previously unseen time-series during training. The Poly class of models train on many synthetic LPPLS time-series augmented with various noise structures in a supervised manner. Given enough training examples, the P-LNN models also outperform state-of-the-art techniques for estimating the parameters of the LPPLS model as evidenced by the comprehensive distribution of parameter errors. Additionally, this class of models is shown to substantially reduce the time to obtain parameter estimates. Finally, we present applications to the diagnostic and prediction of two financial bubble peaks (followed by their crash) and of a famous rockslide. These contributions provide a bridge between deep learning and the study of the prediction of transition times in complex time series.