Time Series Analysis and Forecasting

Modeling and Analyzing Neural Signals with Phase Variability using the Fisher-Rao Registration

This study compares Fisher-Rao Registration and Dynamic Time Warping for analyzing neural signals, showing FRR's superiority in framework, properties, and efficiency, and applies it to fMRI and EEG data.

Lag Penalized Weighted Correlation for Time Series Clustering

The paper introduces a time series clustering method called Lag Penalized Weighted Correlation (LPWC) that captures temporal patterns and outperforms existing methods in simulated and biological datasets.

PRINCIPAL COMPONENT ANALYSIS AND ITS GENERALIZATIONSFOR ANY TYPE SEQUENCE (PCA-SEQ)

This paper extends singular spectral analysis (SSA) to any sequence type, introducing PCA-Seq for analyzing non-numerical data like molecular sequences.

catch22: CAnonical Time-series CHaracteristics

The paper introduces catch22, a set of 22 time-series features that balance computational efficiency and classification performance across diverse biomedical and other time-series datasets.

Automatic time-series phenotyping using massive feature extraction

This paper introduces hctsa, a tool that automatically extracts interpretable time-series features to relate complex time-series data to scientific outcomes.