Machine Learning Advances for Time Series Forecasting (2020-12-23T00:00:00.000000Z)

TL;DR

The most recent advances in supervised machine learning and highdimensional models for time series forecasting are surveyed and ensemble and hybrid models by combining ingredients from different alternatives are considered.

Abstract

In this paper we survey the most recent advances in supervised machine learning and high-dimensional models for time series forecasting. We consider both linear and nonlinear alternatives. Among the linear methods we pay special attention to penalized regressions and ensemble of models. The nonlinear methods considered in the paper include shallow and deep neural networks, in their feed-forward and recurrent versions, and tree-based methods, such as random forests and boosted trees. We also consider ensemble and hybrid models by combining ingredients from different alternatives. Tests for superior predictive ability are briefly reviewed. Finally, we discuss application of machine learning in economics and finance and provide an illustration with high-frequency financial data.

Authors

Ricardo P. Masini

1 papers

M. C. Medeiros

1 papers

Eduardo F. Mendes

1 papers

TL;DR

Abstract

Authors

References169 items

On the Aggregation of Probability Assessments: Regularized Mixtures of Predictive Densities for Eurozone Inﬂation and Real Interest Rates

Machine Learning Panel Data Regressions with an Application to Nowcasting Price Earnings Ratios

Lasso inference for high-dimensional time series

Diverging Tests of Equal Predictive Ability

Targeting predictors in random forest regression

A Note on Adaptive Group Lasso for Structural Break Time Series

Conditional Superior Predictive Ability

Modified LASSO estimators for time series regression models with dependent disturbances

Nonconcave penalized estimation in sparse vector autoregression model

Modeling of time series using random forests: theoretical developments

Regularized estimation of high‐dimensional vector autoregressions with weakly dependent innovations

Inference for High-Dimensional Regressions With Heteroskedasticity and Auto-correlation

Machine Learning Time Series Regressions With an Application to Nowcasting

How is Machine Learning Useful for Macroeconomic Forecasting?

In Search of a Job: Forecasting Employment Growth Using Google Trends

Forecasting Inflation in a Data-Rich Environment: The Benefits of Machine Learning Methods

Forecasting U.S. money growth using economic uncertainty measures and regularisation techniques

Granger Causality Testing in High-Dimensional VARs: a Post-Double-Selection Procedure

Empirical Asset Pricing Via Machine Learning

On LASSO for predictive regression

Deep Neural Networks for Estimation and Inference: Application to Causal Effects and Other Semiparametric Estimands

An automated approach towards sparse single-equation cointegration modelling

Machine Learning for Regularized Survey Forecast Combination: Partially-Egalitarian Lasso and its Derivatives

Macroeconomic forecasting using penalized regression methods

Lasso guarantees for $\beta$-mixing heavy-tailed time series

Real-time inflation forecasting with high-dimensional models: The case of Brazil

High-dimensional Linear Regression for Dependent Data with Applications to Nowcasting

Lasso for sparse linear regression with exponentially β-mixing errors

Double/Debiased Machine Learning for Treatment and Structural Parameters

Questioning the News About Economic Growth: Sparse Forecasting Using Thousands of News-Based Sentiment Values

Variable Screening for High Dimensional Time Series

Sparse Signals in the Cross-Section of Returns

When and Why Are Deep Networks Better Than Shallow Ones?

Double/Debiased/Neyman Machine Learning of Treatment Effects

Forecasting Macroeconomic Variables Using Neural Network Models and Three Automated Model Selection Techniques

LASSO‐Type Penalties for Covariate Selection and Forecasting in Time Series

High-Dimensional Predictive Regression in the Presence of Cointegration

Error bounds for approximations with deep ReLU networks

Oracle inequalities for high-dimensional prediction

Forecasting in Economics and Finance

ℓ1-regularization of high-dimensional time-series models with non-Gaussian and heteroskedastic errors

Estimating Structured Vector Autoregressive Models

Nonlinearity, Breaks, and Long-Range Dependence in Time-Series Models

Estimation and Inference of Heterogeneous Treatment Effects using Random Forests

CONSISTENT AND CONSERVATIVE MODEL SELECTION WITH THE ADAPTIVE LASSO IN STATIONARY AND NONSTATIONARY AUTOREGRESSIONS

Adaptive LASSO estimation for ARDL models with GARCH innovations

Statistical Learning with Sparsity: The Lasso and Generalizations

Complete subset regressions with large-dimensional sets of predictors

A fast unified algorithm for solving group-lasso penalize learning problems

Forecasting performances of three automated modelling techniques during the economic crisis 2007–2009

Penalized Estimation of Semi-Parametric Additive Time-Series Models

Consistency of Random Forests

Post-selection adaptive inference for Least Angle Regression and the Lasso

Complete subset regressions

Exact post-selection inference, with application to the lasso

Regularized estimation in sparse high-dimensional time series models

Lassoing the HAR Model: A Model Selection Perspective on Realized Volatility Dynamics

Penalized regression models with autoregressive error terms

Lag weighted lasso for time series model

On asymptotically optimal confidence regions and tests for high-dimensional models

A SIGNIFICANCE TEST FOR THE LASSO.

A Sparse-Group Lasso

STRONG ORACLE OPTIMALITY OF FOLDED CONCAVE PENALIZED ESTIMATION.

Comparing Predictive Accuracy, Twenty Years Later: A Personal Perspective on the Use and Abuse of Diebold–Mariano Tests

Oracle Inequalities for High Dimensional Vector Autoregressions

Oracle Efficient Estimation and Forecasting with the Adaptive Lasso and the Adaptive Group Lasso in Vector Autoregressions

Inference on Treatment Effects after Selection Amongst High-Dimensional Controls

Simultaneous sparse model selection and coefficient estimation for heavy-tailed autoregressive processes

Modelling nonlinear economic time series

Subset selection for vector autoregressive processes via adaptive Lasso

The Benefits of Bagging for Forecast Models of Realized Volatility

The Model Confidence Set

Bayesian lasso regression

ON THE ADAPTIVE ELASTIC-NET WITH A DIVERGING NUMBER OF PARAMETERS.

Asymmetric effects and long memory in the volatility of Dow Jones stocks

Smoothly Clipped Absolute Deviation on High Dimensions