High-Dimensional Multivariate Forecasting with Low-Rank Gaussian Copula Processes

Predicting the dependencies between observations from multiple time series is critical for applications such as anomaly detection, financial risk management, causal analysis, or demand forecasting. However, the computational and numerical difficulties of estimating time-varying and high-dimensional covariance matrices often limits existing methods to handling at most a few hundred dimensions or requires making strong assumptions on the dependence between series. We propose to combine an RNN-based time series model with a Gaussian copula process output model with a low-rank covariance structure to reduce the computational complexity and handle non-Gaussian marginal distributions. This permits to drastically reduce the number of parameters and consequently allows the modeling of time-varying correlations of thousands of time series. We show on several real-world datasets that our method provides significant accuracy improvements over state-of-the-art baselines and perform an ablation study analyzing the contributions of the different components of our model.

[1]  Andrew Gordon Wilson,et al.  Copula Processes , 2010, NIPS.

[2]  K. Torkkola,et al.  A Multi-Horizon Quantile Recurrent Forecaster , 2017, 1711.11053.

[3]  Heiga Zen,et al.  WaveNet: A Generative Model for Raw Audio , 2016, SSW.

[4]  Larry A. Wasserman,et al.  High Dimensional Semiparametric Gaussian Copula Graphical Models. , 2012, ICML 2012.

[5]  Guokun Lai,et al.  Modeling Long- and Short-Term Temporal Patterns with Deep Neural Networks , 2017, SIGIR.

[6]  Helmut Ltkepohl,et al.  New Introduction to Multiple Time Series Analysis , 2007 .

[7]  A. Kock,et al.  Oracle Inequalities for High Dimensional Vector Autoregressions , 2012, 1311.0811.

[8]  Quoc V. Le,et al.  Sequence to Sequence Learning with Neural Networks , 2014, NIPS.

[9]  Jean Boivin,et al.  Measuring the Effects of Monetary Policy: A Factor-Augmented Vector Autoregressive (FAVAR) Approach , 2003 .

[10]  Jean-François Toubeau,et al.  Deep Learning-Based Multivariate Probabilistic Forecasting for Short-Term Scheduling in Power Markets , 2019, IEEE Transactions on Power Systems.

[11]  Jürgen Schmidhuber,et al.  Long Short-Term Memory , 1997, Neural Computation.

[12]  Ruofeng Wen,et al.  Deep Generative Quantile-Copula Models for Probabilistic Forecasting , 2019, ArXiv.

[13]  Zoubin Ghahramani,et al.  A Unifying Review of Linear Gaussian Models , 1999, Neural Computation.

[14]  Trevor Hastie,et al.  Regularization Paths for Generalized Linear Models via Coordinate Descent. , 2010, Journal of statistical software.

[15]  Wm. R. Wright General Intelligence, Objectively Determined and Measured. , 1905 .

[16]  Simon A. Broda,et al.  Financial Valuation and Risk Management Working Paper No . 454 CHICAGO : A Fast and Accurate Method for Portfolio Risk Calculation , 2008 .

[17]  Larry A. Wasserman,et al.  The Nonparanormal: Semiparametric Estimation of High Dimensional Undirected Graphs , 2009, J. Mach. Learn. Res..

[18]  Melvin J. Hinich,et al.  Time Series Analysis by State Space Methods , 2001 .

[19]  Michael E. Tipping,et al.  Probabilistic Principal Component Analysis , 1999 .

[20]  Yoshua Bengio,et al.  Zoneout: Regularizing RNNs by Randomly Preserving Hidden Activations , 2016, ICLR.

[21]  Hsiang-Fu Yu,et al.  Think Globally, Act Locally: A Deep Neural Network Approach to High-Dimensional Time Series Forecasting , 2019, NeurIPS.

[22]  Kun Zhang,et al.  Efficient factor GARCH models and factor-DCC models , 2009 .

[23]  Barnabás Póczos,et al.  Transformation Autoregressive Networks , 2018, ICML.

[24]  Gal Elidan,et al.  Copulas in Machine Learning , 2013 .

[25]  Neil D. Lawrence,et al.  Auto-Differentiating Linear Algebra , 2017, ArXiv.

[26]  Syama Sundar Rangapuram,et al.  Probabilistic Forecasting with Spline Quantile Function RNNs , 2019, AISTATS.

[27]  Andrew J. Patton A review of copula models for economic time series , 2012, J. Multivar. Anal..

[28]  Iain Murray,et al.  Masked Autoregressive Flow for Density Estimation , 2017, NIPS.

[29]  Mark W. Watson,et al.  Dynamic Factor Models, Factor-Augmented Vector Autoregressions, and Structural Vector Autoregressions in Macroeconomics , 2016 .

[30]  Syama Sundar Rangapuram,et al.  GluonTS: Probabilistic Time Series Models in Python , 2019, ArXiv.

[31]  R. L. Winkler,et al.  Scoring Rules for Continuous Probability Distributions , 1976 .

[32]  D. Harville Matrix Algebra From a Statistician's Perspective , 1998 .

[33]  Laurent A.F. Callot,et al.  Oracle Efficient Estimation and Forecasting with the Adaptive Lasso and the Adaptive Group Lasso in Vector Autoregressions , 2012 .

[34]  D. Cox,et al.  An Analysis of Transformations , 1964 .

[35]  Roy van der Weide,et al.  GO-GARCH: a multivariate generalized orthogonal GARCH model , 2002 .

[36]  A. Raftery,et al.  Strictly Proper Scoring Rules, Prediction, and Estimation , 2007 .

[37]  B. Everitt,et al.  An Introduction to Latent Variable Models , 1984 .

[38]  M. Sklar Fonctions de repartition a n dimensions et leurs marges , 1959 .

[39]  L. Bauwens,et al.  Multivariate GARCH Models: A Survey , 2003 .

[40]  Brian Everitt,et al.  An Introduction to Latent Variable Models , 1984 .

[41]  Wolfgang Aussenegg,et al.  A new copula approach for high-dimensional real world portfolios , 2012 .

[42]  R. Engle Dynamic Conditional Correlation , 2002 .

[43]  Dorothy T. Thayer,et al.  EM algorithms for ML factor analysis , 1982 .

[44]  Alex Graves,et al.  Generating Sequences With Recurrent Neural Networks , 2013, ArXiv.

[45]  Alex Graves,et al.  DRAW: A Recurrent Neural Network For Image Generation , 2015, ICML.

[46]  Valentin Flunkert,et al.  DeepAR: Probabilistic Forecasting with Autoregressive Recurrent Networks , 2017, International Journal of Forecasting.