GRATIS: GeneRAting TIme Series with diverse and controllable characteristics

Abstract

The explosion of time series data in recent years has brought a flourish of new time series analysis methods, for forecasting, clustering, classification and other tasks. The evaluation of these new methods requires either collecting or simulating a diverse set of time series benchmarking data to enable reliable comparisons against alternative approaches. We propose GeneRAting TIme Series with diverse and controllable characteristics, named GRATIS, with the use of mixture autoregressive (MAR) models. We simulate sets of time series using MAR models and investigate the diversity and coverage of the generated time series in a time series feature space. By tuning the parameters of the MAR models, GRATIS is also able to efficiently generate new time series with controllable features. In general, as a costless surrogate to the traditional data collection approach, GRATIS can be used as an evaluation tool for tasks such as time series forecasting and classification. We illustrate the usefulness of our time series generation process through a time series forecasting application.

References in corpus (3)

• Deep learning for time series classification: a review
• Highly comparative feature-based time-series classification
• Approximation of conditional densities by smooth mixtures of regressions

Cited by in corpus (20)

• An Empirical Survey of Data Augmentation for Time Series Classification with Neural Networks
• Forecasting with time series imaging
• Distributed ARIMA Models for Ultra-long Time Series
• What is the best RNN-cell structure to forecast each time series behavior?
• FFORMPP: Feature-based forecast model performance prediction
• Global-scale massive feature extraction from monthly hydroclimatic time series: Statistical characterizations, spatial patterns and hydrological similarity
• Forecasting: theory and practice
• Controllable Data Generation by Deep Learning: A Review
• Role of Data Augmentation Strategies in Knowledge Distillation for Wearable Sensor Data
• Massive feature extraction for explaining and foretelling hydroclimatic time series forecastability at the global scale
• A systematic data characteristic understanding framework towards physical-sensor big data challenges
• Déjà vu: A data-centric forecasting approach through time series cross-similarity
• Fractal interpolation in the context of prediction accuracy optimization
• Hydroclimatic time series features at multiple time scales
• Features of the Earth's seasonal hydroclimate: Characterizations and comparisons across the Koppen-Geiger climates and across continents
• RHiOTS: A Framework for Evaluating Hierarchical Time Series Forecasting Algorithms
• A review of machine learning concepts and methods for addressing challenges in probabilistic hydrological post-processing and forecasting
• Forecast with Forecasts: Diversity Matters
• Personalized Online Machine Learning
• Time series features for supporting hydrometeorological explorations and predictions in ungauged locations using large datasets