scholarly journals Performance evaluation of deep neural networks for forecasting time‐series with multiple structural breaks and high volatility

2021 ◽  
Author(s):  
Rohit Kaushik ◽  
Shikhar Jain ◽  
Siddhant Jain ◽  
Tirtharaj Dash
Keyword(s):  
Neural Networks ◽  
Time Series ◽  
Performance Evaluation ◽  
Structural Breaks ◽  
Deep Neural Networks ◽  
Multiple Structural Breaks ◽  
High Volatility

scholarly journals Forecasting Reservoir Water Levels Using Deep Neural Networks: A Case Study of Angat Dam in the Philippines

Water ◽  
2021 ◽  
Vol 14 (1) ◽  
pp. 34
Author(s):  
Sebastian C. Ibañez ◽  
Carlo Vincienzo G. Dajac ◽  
Marissa P. Liponhay ◽  
Erika Fille T. Legara ◽  
Jon Michael H. Esteban ◽  
...  
Keyword(s):  
Neural Networks ◽  
Time Series ◽  
Deep Neural Networks ◽  
Cross Validation ◽  
The Philippines ◽  
Water Levels ◽  
Short Term ◽  
Exogenous Variables ◽  
Reservoir Water ◽  
Forecasting Performance

Forecasting reservoir water levels is essential in water supply management, impacting both operations and intervention strategies. This paper examines the short-term and long-term forecasting performance of several statistical and machine learning-based methods for predicting the water levels of the Angat Dam in the Philippines. A total of six forecasting methods are compared: naïve/persistence; seasonal mean; autoregressive integrated moving average (ARIMA); gradient boosting machines (GBM); and two deep neural networks (DNN) using a long short-term memory-based (LSTM) encoder-decoder architecture: a univariate model (DNN-U) and a multivariate model (DNN-M). Daily historical water levels from 2001 to 2021 are used in predicting future water levels. In addition, we include meteorological data (rainfall and the Oceanic Niño Index) and irrigation data as exogenous variables. To evaluate the forecast accuracy of our methods, we use a time series cross-validation approach to establish a more robust estimate of the error statistics. Our results show that our DNN-U model has the best accuracy in the 1-day-ahead scenario with a mean absolute error (MAE) and root mean square error (RMSE) of 0.2 m. In the 30-day-, 90-day-, and 180-day-ahead scenarios, the DNN-M shows the best performance with MAE (RMSE) scores of 2.9 (3.3), 5.1 (6.0), and 6.7 (8.1) meters, respectively. Additionally, we demonstrate that further improvements in performance are possible by scanning over all possible combinations of the exogenous variables and only using a subset of them as features. In summary, we provide a comprehensive framework for evaluating water level forecasting by defining a baseline accuracy, analyzing performance across multiple prediction horizons, using time series cross-validation to assess accuracy and uncertainty, and examining the effects of exogenous variables on forecasting performance. In the process, our work addresses several notable gaps in the methodologies of previous works.

A Blockwise Embedding for Multi-Day-Ahead Prediction of Energy Time Series by Randomized Deep Neural Networks

2021 ◽  
Author(s):  
Francesco Di Luzio ◽  
Antonello Rosato ◽  
Federico Succetti ◽  
Massimo Panella
Keyword(s):  
Neural Networks ◽  
Time Series ◽  
Deep Neural Networks

Deep Neural Networks for Mapping Integrated Crop-Livestock Systems Using Planetscope Time Series

2021 ◽  
Author(s):  
Henrique S. L. Almeida ◽  
Aliny A. Dos Reis ◽  
Joao P. S. Werner ◽  
Joao F. G. Antunes ◽  
Liheng Zhong ◽  
...  
Keyword(s):  
Neural Networks ◽  
Time Series ◽  
Deep Neural Networks ◽  
Livestock Systems

Evolving deep neural networks for Time Series Forecasting

2021 ◽  
Vol 18 (2) ◽  
pp. 40-55
Author(s):  
Lídio Mauro Lima Campos ◽  
◽  
Jherson Haryson Almeida Pereira ◽  
Danilo Souza Duarte ◽  
Roberto Célio Limão Oliveira ◽  
...  
Keyword(s):  
Genetic Algorithm ◽  
Neural Networks ◽  
Time Series ◽  
Deep Neural Networks ◽  
Good Prediction ◽  
Biologically Inspired ◽  
Price Prediction ◽  
Error Measures ◽  
Good Ability ◽  
Proposed Model

The aim of this paper is to introduce a biologically inspired approach that can automatically generate Deep Neural networks with good prediction capacity, smaller error and large tolerance to noises. In order to do this, three biological paradigms were used: Genetic Algorithm (GA), Lindenmayer System and Neural Networks (DNNs). The final sections of the paper present some experiments aimed at investigating the possibilities of the method in the forecast the price of energy in the Brazilian market. The proposed model considers a multi-step ahead price prediction (12, 24, and 36 weeks ahead). The results for MLP and LSTM networks show a good ability to predict peaks and satisfactory accuracy according to error measures comparing with other methods.

scholarly journals Transfer Learning for Clinical Time Series Analysis Using Deep Neural Networks

2019 ◽  
Vol 4 (2) ◽  
pp. 112-137 ◽  
Author(s):  
Priyanka Gupta ◽  
Pankaj Malhotra ◽  
Jyoti Narwariya ◽  
Lovekesh Vig ◽  
Gautam Shroff
Keyword(s):  
Neural Networks ◽  
Time Series ◽  
Time Series Analysis ◽  
Transfer Learning ◽  
Deep Neural Networks ◽  
Series Analysis

scholarly journals Deep Learning for Nonlinear Time Series: Examples for Inferring Slow Driving Forces

2020 ◽  
Vol 30 (15) ◽  
pp. 2050226
Author(s):  
Yoshito Hirata ◽  
Kazuyuki Aihara
Keyword(s):  
Neural Networks ◽  
Time Series ◽  
Driving Force ◽  
Deep Neural Networks ◽  
Driving Forces ◽  
Distance Matrix ◽  
Recurrence Plot ◽  
Static Images ◽  
Dynamical Properties ◽  
Delay Coordinates

Records for observing dynamics are usually complied by a form of time series. However, time series can be a challenging type of dataset for deep neural networks to learn. In deep neural networks, pairs of inputs and outputs are usually fed for constructive mapping. Such inputs are typically prepared as static images in successful applications. And so, here we propose two methods to prepare such inputs for learning the dynamical properties behind time series. In the first method, we simply array a time series in the shape of a rectangle as an image. In the second method, we convert a time series into a distance matrix using delay coordinates, or an unthresholded recurrence plot. We demonstrate that the second method performs well in inferring a slow driving force from observations of a forced system within which there are symmetry and almost invariant subsets.

Sign in / Sign up

Export Citation Format

Share Document