scholarly journals Improving Daily Peak Flow Forecasts Using Hybrid Fourier-Series Autoregressive Integrated Moving Average and Recurrent Artificial Neural Network Models

AI ◽  
2020 ◽  
Vol 1 (2) ◽  
pp. 263-275 ◽  
Author(s):  
Mohammad Ebrahim Banihabib ◽  
Reihaneh Bandari ◽  
Mohammad Valipour
Keyword(s):  
Neural Network ◽  
Time Series ◽  
Artificial Neural Network ◽  
Hybrid Model ◽  
Peak Flow ◽  
Moving Average ◽  
Flow Time ◽  
Autoregressive Integrated Moving Average ◽  
Auto Regressive ◽  
Reservoir Inflow

In multi-purpose reservoirs, to achieve optimal operation, sophisticated models are required to forecast reservoir inflow in both short- and long-horizon times with an acceptable accuracy, particularly for peak flows. In this study, an auto-regressive hybrid model is proposed for long-horizon forecasting of daily reservoir inflow. The model is examined for a one-year horizon forecasting of high-oscillated daily flow time series. First, a Fourier-Series Filtered Autoregressive Integrated Moving Average (FSF-ARIMA) model is applied to forecast linear behavior of daily flow time series. Second, a Recurrent Artificial Neural Network (RANN) model is utilized to forecast FSF-ARIMA model’s residuals. The hybrid model follows the detail of observed flow time variation and forecasted peak flow more accurately than previous models. The proposed model enhances the ability to forecast reservoir inflow, especially in peak flows, compared to previous linear and nonlinear auto-regressive models. The hybrid model has a potential to decrease maximum and average forecasting error by 81% and 80%, respectively. The results of this investigation are useful for stakeholders and water resources managers to schedule optimum operation of multi-purpose reservoirs in controlling floods and generating hydropower.

The Forecasting of the Elevator Traffic Flow Time Series Based on ARIMA and GP

2012 ◽  
Vol 588-589 ◽  
pp. 1466-1471 ◽  
Author(s):  
Jun Fang Li ◽  
Qun Zong
Keyword(s):  
Time Series ◽  
Traffic Flow ◽  
Linear Models ◽  
Moving Average ◽  
Arima Model ◽  
Flow Time ◽  
Forecasting Model ◽  
Autoregressive Integrated Moving Average ◽  
Hybrid Forecasting ◽  
Artificial Neural Network Ann

As one of the conventional statistical methods, the autoregressive integrated moving average (ARIMA) model has been one of the most widely used linear models in time series forecasting. However, the ARIMA model cannot easily capture the nonlinear patterns. Artificial neural network (ANN) can be utilized to construct more accurate forecasting model than ARIMA for nonlinear time series, but it is difficult to explain the meaning of the hidden layers of ANN and it does not produce a mathematical equation. In this study, by combining ARIMA with genetic programming (GP), a hybrid forecasting model will be used for elevator traffic flow time series which can improve the accuracy both the GP and the ARIMA forecasting models separately. At last, simulations are adopted to demonstrate the advantages of the proposed ARIMA-GP forecasting model.

scholarly journals STUDI KOMPARASI PREDIKSI CURAH HUJAN METODE FAST FOURIER TRANSFORMATION (FFT), AUTOREGRESSIVE INTEGRATED MOVING AVERAGE (ARIMA) DAN ARTIFICIAL NEURAL NETWORK (ANN)

2015 ◽  
Vol 35 (02) ◽  
pp. 241
Author(s):  
Dyah Susilokarti ◽  
Sigit Supadmo Arif ◽  
Sahid Susanto ◽  
Lilik Sutiarso
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Time Series Data ◽  
Moving Average ◽  
Series Data ◽  
Fast Fourier Transformation ◽  
Autoregressive Integrated Moving Average ◽  
Precipitation Prediction ◽  
Artificial Neural ◽  
Artificial Neural Network Ann

Optimum climate condition and water availability are essential to support strategic venue and time for plants to grow and produce.  Precipitation prediction is needed to determine how much precipitation will provide water for plants on each stage of growth. Nowadays, the high variability of precipitation calls for a prediction model that will accurately foreseethe precipitation condition in the future. The prediction conducted is based on time-series data analysis. The research aims to comparethe effectiveness of three precipitation prediction methods, which are Fast Forier Transformation (FFT), Autoregressive Integrated Moving Average (ARIMA) and Artificial Neural Network (ANN).  Their respective performances are determined by their Mean Square Error (MSE) values.  Methods with highest correlation values and lowest MSE shows the best performance. The MSE result for FFT is 14,92; ARIMA is 17,49; and  ANN is 0,07. This research concluded that Artificial Neural Network (ANN) method showed best performance compare to the other two because it had produced a prediction with the lowest MSE value.Keywords: Precipitation prediction, Fast Forier Transformation (FFT), Autoregressive Integrated Moving Average ABSTRAKKondisi iklim dan ketersediaan air yang optimal bagi pertumbuhan dan perkembangan tanaman sangat diperlukan dalam upaya mendukung strategi budidaya tanaman sesuai ruang dan waktu. Prediksi curah hujan sangat diperlukan untuk untuk mengetahui sejauh mana curah hujan dapat memenuhi kebutuhan air pada setiap tahap pertumbuhantanaman. Variabilitas curah hujan yang tinggi saat ini, membutuhkan pemodelan yang dapat memprediksi secara akurat bagaimana kondisi curah hujan dimasa yang akan datang. Prediksi yang dilakukan adalah prediksi berdasarkan urutan waktu ().  Tujuan dari penelitian ini adalah untuk membandingkan akurasi prediksi curah hujan antara metode  (FFT),  (ARIMA) dan (ANN). Kinerja ketiga metode yang digunakan dilihat dari nilai  (MSE). Metode dengan nilai korelasi tertinggi dan nilai MSE terkecil menunjukkan kinerja terbaik. Hasil penelitan untuk FFT diperoleh nilai MSE = 14,92, ARIMA = 17,49 sedangkan ANN = 0,07. Ini menunjukkan bahwa metode   (ANN) menunjukkan kinerja yang paling baik diantara dua metode lainnya karena menghasilkan prediksi yangmempunyai nilai MSE terkecil.Kata kunci: Prediksi curah hujan,FFT, ARIMA dan ANN 

scholarly journals Konstruksi Forecasting System Multi-Model untuk pemodelan matematika pada peramalan Indeks Pembangunan Manusia Provinsi Nusa Tenggara Barat

2018 ◽  
Vol 4 (2) ◽  
pp. 114 ◽  
Author(s):  
Lalu Sucipto ◽  
Syaharuddin Syaharuddin
Keyword(s):  
Neural Network ◽  
Time Series ◽  
Artificial Neural Network ◽  
Graphical User Interface ◽  
Interpolation Method ◽  
Moving Average ◽  
Back Propagation ◽  
Naive Method ◽  
Forecasting System ◽  
Exponential Smoothing Method

Penelitian ini bertujuan untuk mengembangkan produk Forecasting System Multi-Model (FSM) guna menentukan metode terbaik dalam sistem peramalan (forecast) dengan mengkonstruksi beberapa metode dalam bentuk Graphical User Interface (GUI) Matlab dengan menghitung semua indikator tingkat akurasi guna menemukan model matematika terbaik dari data time series pada periode tertentu. Pada tahap simulasi, tim peneliti menggunakan data Indeks Pembangunan Manusia (IPM) Provinsi Nusa Tenggara Barat (NTB) tahun 2010-2017 guna memprediksi IPM NTB tahun 2018. Adapun metode yang diuji adalah Moving Average (SMA, WMA dan EMA), Exponential Smoothing Method (SES, Brown, Holt, dan Winter), Naive Method, Interpolation Method (Newton Gregory), dan Artificial Neural Network (Back Propagation). Kemudian model dievaluasi untuk melihat tingkat akurasi masing-masing metode berdasarkan nilai MAD, MSE, dan MAPE. Berdasarkan hasil simulasi data dari 10 metode yang diuji diketahui bahwa metode Holt paling akurat dengan hasil prediksi tahun 2018 sebesar 67,45  dengan MAD, MSE, dan MAPE berturut-turut sebesar 0,22654; 0,075955 dan 0,34829.   The purpose of this research is to develop a product was called Forecasting System Multi-Model (FSM) to determine the best method in the forecasting system by constructing several methods in the form of Graphical User Interface (GUI) Matlab. It was done by all indicator accuration to find the best mathematical model of time series data in a certain period. In the simulation phase, this research used the Human Development Index (HDI) data of West Nusa Tenggara (NTB) Province in 2010 - 2017 to predict the HDI data of NTB in 2018. The methods tested were Moving Average (SMA, WMA and EMA), Exponential Smoothing Method (SES, Brown, Holt, and Winter), Naive Method, Interpolation Method (Newton Gregory), and Artificial Neural Network (Back Propagation). Then the models/methods were evaluated to see the level of accuracy of each method based on the value of MAD, MSE, and MAPE. Based on data simulation result from 10 tested method known that Holt method is most accurate with prediction result of 2018 equal to 67,45 with MAD, MSE, and MAPE respectively equal to 0.22654, 0.075955 and 0.34829.

scholarly journals Time series prediction of under-five mortality rates for Nigeria: comparative analysis of artificial neural networks, Holt-Winters exponential smoothing and autoregressive integrated moving average models

2020 ◽  
Vol 20 (1) ◽  
Author(s):  
Daniel Adedayo Adeyinka ◽  
Nazeem Muhajarine
Keyword(s):  
Neural Network ◽  
Time Series ◽  
Mean Absolute Error ◽  
Moving Average ◽  
Absolute Error ◽  
Mortality Rates ◽  
Childhood Mortality ◽  
Autoregressive Integrated Moving Average ◽  
Under Five ◽  
Deming Regression

Abstract Background Accurate forecasting model for under-five mortality rate (U5MR) is essential for policy actions and planning. While studies have used traditional time series modeling techniques (e.g., autoregressive integrated moving average (ARIMA) and Holt-Winters smoothing exponential methods), their appropriateness to predict noisy and non-linear data (such as childhood mortality) has been debated. The objective of this study was to model long-term U5MR with group method of data handling (GMDH)-type artificial neural network (ANN), and compare the forecasts with the commonly used conventional statistical methods—ARIMA regression and Holt-Winters exponential smoothing models. Methods The historical dataset of annual U5MR in Nigeria from 1964 to 2017 was obtained from the official website of World Bank. The optimal models for each forecasting methods were used for forecasting mortality rates to 2030 (ending of Sustainable Development Goal era). The predictive performances of the three methods were evaluated, based on root mean squared errors (RMSE), root mean absolute error (RMAE) and modified Nash-Sutcliffe efficiency (NSE) coefficient. Statistically significant differences in loss function between forecasts of GMDH-type ANN model compared to each of the ARIMA and Holt-Winters models were assessed with Diebold-Mariano (DM) test and Deming regression. Results The modified NSE coefficient was slightly lower for Holt-Winters methods (96.7%), compared to GMDH-type ANN (99.8%) and ARIMA (99.6%). The RMSE of GMDH-type ANN (0.09) was lower than ARIMA (0.23) and Holt-Winters (2.87). Similarly, RMAE was lowest for GMDH-type ANN (0.25), compared with ARIMA (0.41) and Holt-Winters (1.20). From the DM test, the mean absolute error (MAE) was significantly lower for GMDH-type ANN, compared with ARIMA (difference = 0.11, p-value = 0.0003), and Holt-Winters model (difference = 0.62, p-value< 0.001). Based on the intercepts from Deming regression, the predictions from GMDH-type ANN were more accurate (β0 = 0.004 ± standard error: 0.06; 95% confidence interval: − 0.113 to 0.122). Conclusions GMDH-type neural network performed better in predicting and forecasting of under-five mortality rates for Nigeria, compared to the ARIMA and Holt-Winters models. Therefore, GMDH-type ANN might be more suitable for data with non-linear or unknown distribution, such as childhood mortality. GMDH-type ANN increases forecasting accuracy of childhood mortalities in order to inform policy actions in Nigeria.

Sign in / Sign up

Export Citation Format

Share Document