Large-scale assessment of Prophet for multi-step ahead forecasting of monthly streamflow

Abstract. We assess the performance of the recently introduced Prophet model in multi-step ahead forecasting of monthly streamflow by using a large dataset. Our aim is to compare the results derived through two different approaches. The first approach uses past information about the time series to be forecasted only (standard approach), while the second approach uses exogenous predictor variables alongside with the use of the endogenous ones. The additional information used in the fitting and forecasting processes includes monthly precipitation and/or temperature time series, and their forecasts respectively. Specifically, the exploited exogenous (observed or forecasted) information considered at each time step exclusively concerns the time of interest. The algorithms based on the Prophet model are in total four. Their forecasts are also compared with those obtained using two classical algorithms and two benchmarks. The comparison is performed in terms of four metrics. The findings suggest that the compared approaches are equally useful.

[1]  Martyn P. Clark,et al.  Development of a large-sample watershed-scale hydrometeorological data set for the contiguous USA: data set characteristics and assessment of regional variability in hydrologic model performance , 2014 .

[2]  Demetris Koutsoyiannis,et al.  A Bayesian statistical model for deriving the predictive distribution of hydroclimatic variables , 2014, Climate Dynamics.

[3]  Demetris Koutsoyiannis,et al.  On the long-range dependence properties of annual precipitation using a global network of instrumental measurements , 2018 .

[4]  Demetris Koutsoyiannis,et al.  Medium-range flow prediction for the Nile: a comparison of stochastic and deterministic methods / Prévision du débit du Nil à moyen terme: une comparaison de méthodes stochastiques et déterministes , 2008 .

[5]  Soroosh Sorooshian,et al.  Developing reservoir monthly inflow forecasts using artificial intelligence and climate phenomenon information , 2017 .

[6]  Yihui Xie,et al.  Dynamic Documents with R and knitr , 2015 .

[7]  Hadley Wickham,et al.  Dates and Times Made Easy with lubridate , 2011 .

[8]  Demetris Koutsoyiannis,et al.  Forecasting of geophysical processes using stochastic and machine learning algorithms: Supplementary material , 2017 .

[9]  Demetris Koutsoyiannis,et al.  Error Evolution in Multi-Step Ahead Streamflow Forecasting for the Operation of Hydropower Reservoirs , 2017 .

[10]  Demetris Koutsoyiannis,et al.  Predictability of monthly temperature and precipitation using automatic time series forecasting methods , 2018, Acta Geophysica.

[11]  Yaxing Wei,et al.  Daymet: Daily Surface Weather Data on a 1-km Grid for North America, Version 2 , 2014 .

[12]  Hadley Wickham,et al.  Tools to Make Developing R Packages Easier , 2016 .

[13]  Martyn P. Clark,et al.  The CAMELS data set: catchment attributes and meteorology for large-sample studies , 2017 .

[14]  David H. Wolpert,et al.  The Lack of A Priori Distinctions Between Learning Algorithms , 1996, Neural Computation.

[15]  Demetris Koutsoyiannis,et al.  Simultaneous estimation of the parameters of the Hurst–Kolmogorov stochastic process , 2011 .

[16]  A. Zeileis,et al.  zoo: S3 Infrastructure for Regular and Irregular Time Series , 2005, math/0505527.

[17]  Rob J Hyndman,et al.  Automatic Time Series Forecasting: The forecast Package for R , 2008 .

[18]  Georgia Papacharalampous,et al.  Variable Selection in Time Series Forecasting Using Random Forests , 2017, Algorithms.

[19]  S. Bagli,et al.  Operational River Discharge Forecasting with Support Vector Regression Technique Applied to Alpine Catchments: Results, Advantages, Limits and Lesson Learned , 2017, Water Resources Management.

[20]  Yihui Xie,et al.  knitr: A Comprehensive Tool for Reproducible Research in R , 2018, Implementing Reproducible Research.

[21]  Demetris Koutsoyiannis,et al.  Comparison of stochastic and machine learning methods for multi-step ahead forecasting of hydrological processes , 2019, Stochastic Environmental Research and Risk Assessment.

[22]  P. Krause,et al.  COMPARISON OF DIFFERENT EFFICIENCY CRITERIA FOR HYDROLOGICAL MODEL ASSESSMENT , 2005 .

[23]  Rosangela Ballini,et al.  Multi-step-ahead monthly streamflow forecasting by a neurofuzzy network model , 2001, Proceedings Joint 9th IFSA World Congress and 20th NAFIPS International Conference (Cat. No. 01TH8569).

[24]  J. Nash,et al.  River flow forecasting through conceptual models part I — A discussion of principles☆ , 1970 .

[25]  Tao Hong,et al.  Probabilistic electric load forecasting: A tutorial review , 2016 .

[26]  Hoshin Vijai Gupta,et al.  Decomposition of the mean squared error and NSE performance criteria: Implications for improving hydrological modelling , 2009 .

[27]  Alexander Y. Sun,et al.  Monthly streamflow forecasting using Gaussian Process Regression , 2014 .

[28]  Monthly streamflow forecast for National Interconnected System (NIS) using Periodic Auto-regressive Endogenous Models (PAR) and Exogenous (PARX) with climate information , 2017 .

[29]  C. Notarnicola,et al.  Seasonal river discharge forecasting using support vector regression: A case study in the Italian Alps , 2015 .

[30]  Demetris Koutsoyiannis,et al.  One-step ahead forecasting of geophysical processes within a purely statistical framework , 2018, Geoscience Letters.

[31]  Benjamin Letham,et al.  Forecasting at Scale , 2018, PeerJ Prepr..