Method For Time Series Forecasting Research Articles

Learning Future-Aware Correlation Filters for Efficient UAV Tracking

In recent years, discriminative correlation filter (DCF)-based trackers have made considerable progress and drawn widespread attention in the unmanned aerial vehicle (UAV) tracking community. Most existing trackers collect historical information, e.g., training samples, previous filters, and response maps, to promote their discrimination and robustness. Under UAV-specific tracking challenges, e.g., fast motion and view change, variations of both the target and its environment in the new frame are unpredictable. Interfered by future unknown environments, trackers that trained with historical information may be confused by the new context, resulting in tracking failure. In this paper, we propose a novel future-aware correlation filter tracker, i.e., FACF. The proposed method aims at effectively utilizing context information in the new frame for better discriminative and robust abilities, which consists of two stages: future state awareness and future context awareness. In the former stage, an effective time series forecast method is employed to reason a coarse position of the target, which is the reference for obtaining a context patch in the new frame. In the latter stage, we firstly obtain the single context patch with an efficient target-aware method. Then, we train a filter with the future context information in order to perform robust tracking. Extensive experimental results obtained from three UAV benchmarks, i.e., UAV123_10fps, DTB70, and UAVTrack112, demonstrate the effectiveness and robustness of the proposed tracker. Our tracker has comparable performance with other state-of-the-art trackers while running at ∼49 FPS on a single CPU.

EBITDA Index Prediction Using Exponential Smoothing and ARIMA Model

Forecasting has become essential in different economic sectors for decision making in local and regional policies. Therefore, the aim of this paper is to use and compare performance of two linear models to predict future values of a measure of real profit for a group of companies in the fashion sector, as a financial strategy to determine the economic behavior of this industry. With forecasting purposes, Exponential Smoothing (ES) and autoregressive integrated moving averages (ARIMA) models were used for yearly data. ES and ARIMA models are widely used in statistical methods for time series forecasting. Accuracy metrics were used to select the model with best performance and ES parameters. For the real profit measure of the financial performance of the fashion sector in Colombia EBITDA (Earnings Before Interest, Taxes, Depreciation, and Amortization) was used and was calculated using multiple SQL queries.

Evaluation of interpretability methods for multivariate time series forecasting.

Being able to interpret a model’s predictions is a crucial task in many machine learning applications. Specifically, local interpretability is important in determining why a model makes particular predictions. Despite the recent focus on interpretable Artificial Intelligence (AI), there have been few studies on local interpretability methods for time series forecasting, while existing approaches mainly focus on time series classification tasks. In this study, we propose two novel evaluation metrics for time series forecasting: Area Over the Perturbation Curve for Regression and Ablation Percentage Threshold. These two metrics can measure the local fidelity of local explanation methods. We extend the theoretical foundation to collect experimental results on four popular datasets. Both metrics enable a comprehensive comparison of numerous local explanation methods, and an intuitive approach to interpret model predictions. Lastly, we provide heuristical reasoning for this analysis through an extensive numerical study.

Explainable boosted linear regression for time series forecasting

Time series forecasting involves collecting and analyzing past observations to develop a model to extrapolate such observations into the future. Forecasting of future events is important in many fields to support decision making as it contributes to reducing the future uncertainty. We propose explainable boosted linear regression (EBLR) algorithm for time series forecasting, which is an iterative method that starts with a base model, and explains the model’s errors through regression trees. At each iteration, the path leading to highest error is added as a new variable to the base model. In this regard, our approach can be considered as an improvement over general time series models since it enables incorporating nonlinear features by residual explanation. More importantly, use of the single rule that contributes to the error most enables access to interpretable results. The proposed approach extends to probabilistic forecasting through generating prediction intervals based on the empirical error distribution. We conduct a detailed numerical study with EBLR and compare against various other approaches. We observe that EBLR substantially improves the base model performance through extracted features, and provide a comparable performance to other well established approaches. The interpretability of the model predictions and high predictive accuracy of EBLR makes it a promising method for time series forecasting.

On application of machine learning method for history matching and forecasting of times series data from hydrocarbon recovery process using water flooding

The focus of this paper is on application of advance data analytics and deep machine learning methods for time series forecasting of injection/production data from subsurface hydrocarbon recovery processes. Injection and production data, from subsurface reservoir developed though implementation of water flooding recovery mechanism, is used with aim of developing a machine learning based forecasting methodology to eventually replace numerical simulation-based forecasting methods. Different machine learning algorithms exists for single- and multi-step time series forecasting, e.g., nonlinear regression, artificial neural network (ANN) and long-short-term-memory (LSTM) based recurrent neural networks (RNN). In this paper, RNN-LSTM algorithm is tested on real field data, comprising of a number of injection and production well patterns, and results of RNN-LSTM forecasting model are presented. Complexities related to data acquisition, analysis, and processing are also discussed in detail. The results presented in this paper are unique as it is for the first time that ML model is applied to forecast production from a tight carbonate hydrocarbon reservoir, which is developed through water flooding with help of long horizontal wells. Such an application for time series forecasting using RNN-LSTM model has never been presented, before in literature, to the best of author’s knowledge.

Traditional vs. Machine-Learning Methods for Forecasting Sandy Shoreline Evolution Using Historic Satellite-Derived Shorelines

Forecasting shoreline evolution for sandy coasts is important for sustainable coastal management, given the present-day increasing anthropogenic pressures and a changing future climate. Here, we evaluate eight different time-series forecasting methods for predicting future shorelines derived from historic satellite-derived shorelines. Analyzing more than 37,000 transects around the globe, we find that traditional forecast methods altogether with some of the evaluated probabilistic Machine Learning (ML) time-series forecast algorithms, outperform Ordinary Least Squares (OLS) predictions for the majority of the sites. When forecasting seven years ahead, we find that these algorithms generate better predictions than OLS for 54% of the transect sites, producing forecasts with, on average, 29% smaller Mean Squared Error (MSE). Importantly, this advantage is shown to exist over all considered forecast horizons, i.e., from 1 up to 11 years. Although the ML algorithms do not produce significantly better predictions than traditional time-series forecast methods, some proved to be significantly more efficient in terms of computation time. We further provide insight in how these ML algorithms can be improved so that they can be expected to outperform not only OLS regression, but also the traditional time-series forecast methods. These forecasting algorithms can be used by coastal engineers, managers, and scientists to generate future shoreline prediction at a global level and derive conclusions thereof.

Air Pollution Prediction Using an Ensemble of Dynamic Transfer Models for Multivariate Time Series

Entering a new era of big data, analysis of large amounts of real-time data is important, and air quality data as streaming time series are measured by several different sensors. To this end, numerous methods for time-series forecasting and deep-learning approaches based on neural networks have been used. However, they usually rely on a certain model with a stationary condition, and there are few studies of real-time prediction of dynamic massive multivariate data. Use of a variety of independent variables included in the data is important to improve forecasting performance. In this paper, we proposed a real-time prediction approach based on an ensemble method for multivariate time-series data. The suggested method can select multivariate time-series variables and incorporate real-time updatable autoregressive models in terms of performance. We verified the proposed model using simulated data and applied it to predict air quality measured by five sensors and failures based on real-time performance log data in server systems. We found that the proposed method for air pollution prediction showed effective and stable performance for both short- and long-term prediction tasks. In addition, traditional methods for abnormality detection have focused on present status of objects as either normal or abnormal based on provided data, we protectively predict expected statuses of objects with provided real-time data and implement effective system management in cloud environments through the proposed method.

Using a GMDH-type neural network and ARIMA model to forecasting GDP in Algeria during the period of 1990-2019

Forecasting is a method to predict the future using data and the last information as a tool assists in planning to be effective. GMDH-Type (Group Method of Data Handling) artificial neural network (ANN) and Box-Jenkins method are among the know methods for time series forecasting of mathematical modeling. in the present study GMDH-type neural network and ARIMA method has been used to forecasted GDP in Algeria during the period 1990 to2019 (Time series of quarterly observations on Gross Domestic Product (GDP) is used). Root mean square error (RMSE) was used as performance indices to test the accuracy of the forecast. The empirical results for both models showed that the GMDH model is a powerful tool in forecasting GDP and it provides a promising technique in time series forecasting methods.

Short-term Demand Forecasting for Online Car-hailing Services Using Recurrent Neural Networks

ABSTRACT Short-term traffic flow prediction is one of the crucial issues in intelligent transportation system, which is an important part of smart cities. Accurate predictions can enable both the drivers and the passengers to make better decisions about their travel route, departure time, and travel origin selection, which can be helpful in traffic management. Multiple models and algorithms based on time-series prediction and machine learning were applied to this issue and achieved acceptable results. Recently, the availability of sufficient data and computational power motivates us to improve the prediction accuracy via deep-learning approaches. Recurrent neural networks have become one of the most popular methods for time-series forecasting; however, due to the variety of these networks, the question that which type is the most appropriate one for this task remains unsolved. In this paper, we use three kinds of recurrent neural networks including simple RNN units, GRU, and LSTM neural network to predict short-term traffic flow. The dataset from TAP30 Corporation is used for building the models and comparing RNNs with several well-known models, such as DEMA, LASSO, and XGBoost. The results show that all three types of RNNs outperform the others; however, more simple RNNs such as simple recurrent units and GRU perform work better than LSTM in terms of accuracy and training time.

Forecasting of Coalbed Methane Daily Production Based on T-LSTM Neural Networks

Accurately forecasting the daily production of coalbed methane (CBM) is important forformulating associated drainage parameters and evaluating the economic benefit of CBM mining. Daily production of CBM depends on many factors, making it difficult to predict using conventional mathematical models. Because traditional methods do not reflect the long-term time series characteristics of CBM production, this study first used a long short-term memory neural network (LSTM) and transfer learning (TL) method for time series forecasting of CBM daily production. Based on the LSTM model, we introduced the idea of transfer learning and proposed a Transfer-LSTM (T-LSTM) CBM production forecasting model. This approach first uses a large amount of data similar to the target to pretrain the weights of the LSTM network, then uses transfer learning to fine-tune LSTM network parameters a second time, so as to obtain the final T-LSTM model. Experiments were carried out using daily CBM production data for the Panhe Demonstration Zone at southern Qinshui basin in China. Based on the results, the idea of transfer learning can solve the problem of insufficient samples during LSTM training. Prediction results for wells that entered the stable period earlier were more accurate, whereas results for types with unstable production in the early stage require further exploration. Because CBM wells daily production data have symmetrical similarities, which can provide a reference for the prediction of other wells, so our proposed T-LSTM network can achieve good results for the production forecast and can provide guidance for forecasting production of CBM wells.

AI in Healthcare: Time-Series Forecasting Using Statistical, Neural, and Ensemble Architectures.

Both statistical and neural methods have been proposed in the literature to predict healthcare expenditures. However, less attention has been given to comparing predictions from both these methods as well as ensemble approaches in the healthcare domain. The primary objective of this paper was to evaluate different statistical, neural, and ensemble techniques in their ability to predict patients' weekly average expenditures on certain pain medications. Two statistical models, persistence (baseline) and autoregressive integrated moving average (ARIMA), a multilayer perceptron (MLP) model, a long short-term memory (LSTM) model, and an ensemble model combining predictions of the ARIMA, MLP, and LSTM models were calibrated to predict the expenditures on two different pain medications. In the MLP and LSTM models, we compared the influence of shuffling of training data and dropout of certain nodes in MLPs and nodes and recurrent connections in LSTMs in layers during training. Results revealed that the ensemble model outperformed the persistence, ARIMA, MLP, and LSTM models across both pain medications. In general, not shuffling the training data and adding the dropout helped the MLP models and shuffling the training data and not adding the dropout helped the LSTM models across both medications. We highlight the implications of using statistical, neural, and ensemble methods for time-series forecasting of outcomes in the healthcare domain.

Use of Empirical Mode Decomposition in Improving Neural Network Forecasting of Paddy Price

Since rice is a staple food in Malaysia, its price fluctuations pose risks to the producers, suppliers and consumers. Hence, an accurate prediction of paddy price is essential to aid the planning and decision-making in related organizations. The artificial neural network (ANN) has been widely used as a promising method for time series forecasting. In this paper, the effectiveness of integrating empirical mode decomposition (EMD) into an ANN model to forecast paddy price is investigated. The hybrid method is applied on a series of monthly paddy prices fromFebruary 1999 up toMay 2018 as recorded in the Malaysian Ringgit (MYR) per metric tons. The performance of the simple ANN model and the EMD-ANN model was measured and compared based on their root mean squared Error (RMSE), mean absolute error (MAE) and mean percentage error (MPE). This study finds that the integration of EMD into the neural network model improves the forecasting capabilities. The use of EMD in the ANN model made the forecast errors reduced significantly, and the RMSE was reduced by 0.012, MAE by 0.0002 and MPE by 0.0448.

Multi-step time series prediction intervals using neuroevolution

Multi-step time series forecasting (TSF) is a crucial element to support tactical decisions (e.g., designing production or marketing plans several months in advance). While most TSF research addresses only single-point prediction, prediction intervals (PIs) are useful to reduce uncertainty related to important decision making variables. In this paper, we explore a large set of neural network methods for multi-step TSF and that directly optimize PIs. This includes multi-step adaptations of recently proposed PI methods, such as lower–upper bound estimation (LUBET), its ensemble extension (LUBEXT), a multi-objective evolutionary algorithm LUBE (MLUBET) and a two-phase learning multi-objective evolutionary algorithm (M2LUBET). We also explore two new ensemble variants for the evolutionary approaches based on two PI coverage–width split methods (radial slices and clustering), leading to the MLUBEXT, M2LUBEXT, MLUBEXT2 and M2LUBEXT2 methods. A robust comparison was held by considering the rolling window procedure, nine time series from several real-world domains and with different characteristics, two PI quality measures (coverage error and width) and the Wilcoxon statistic. Overall, the best results were achieved by the M2LUBET neuroevolution method, which requires a reasonable computational effort for time series with a few hundreds of observations.

Density-based unsupervised ensemble learning methods for time series forecasting of aggregated or clustered electricity consumption

This paper presents a comparison of the impact of various unsupervised ensemble learning methods on electricity load forecasting. The electricity load from consumers is simply aggregated or optimally clustered to more predictable groups by cluster analysis. The clustering approach consists of efficient preprocessing of data obtained from smart meters by a model-based representation and the K-means method. We have implemented two types of unsupervised ensemble learning methods to investigate the performance of forecasting on clustered or simply aggregated load: bootstrap aggregating based and the newly proposed density-clustering based. Three new bootstrapping methods for time series analysis methods were newly proposed in order to handle the noisy behaviour of time series. The smart meter datasets used in our experiments come from Australia, London, and Ireland, where data from residential consumers were available. The achieved results suggest that for extremely fluctuating and noisy time series the forecasting accuracy improvement through the bagging can be a challenging task. However, our experimental evaluation shows that in most of the cases the density-based unsupervised ensemble learning methods are significantly improving forecasting accuracy of aggregated or clustered electricity load.

Hesitant fuzzy set based computational method for financial time series forecasting

Non-stochastic hesitation in fuzzy time series forecasting methods occurs due to availability of more than one fuzzification methods of time series data. Recently hesitant fuzzy set has gained attention of the researchers to address issue of aforesaid non-stochastic hesitation. In this research paper, we propose and develop a computational algorithm-based method for financial time series forecasting using hesitant fuzzy set. The proposed method uses hesitant fuzzy logical relations that are constructed using triangular fuzzy sets with equal and unequal intervals. In the proposed forecasting method, hesitant fuzzy logical relations are aggregated using a hesitant fuzzy aggregation operator. Advantages of developed hesitant fuzzy set based forecasting method are that it is easy to implement, can cope with huge time series database and enhances the accuracy in financial time series forecast. To see the performance of proposed method in financial time series forecasting, it is implemented on two financial experimental dataset of TAIFEX and SBI share price at BSE. Rigorous comparison analysis of the proposed forecasting method is done by comparing it with other conventional and computational fuzzy time series forecasting methods in terms of RMSE, AFER. Significance of accuracy enhancement in forecasted outputs is also verified using two-tailed t test.

ARIMA model forecast based on EViews software

Time series is a series of data obtained in chronological order. Future values of most time series can be forecasted according to current values and past values. The EViews software is a software package specifically designed to process time series data. Autoregressive Integrated Moving Average (ARIMA) model, a time series forecast method, can be achieved with the EViews software. Based on the EViews software, the forecast procedure with ARIMA model is illustrated in this work. As an example, the Gross Domestic Product (GDP) of China is forecasted from 2016 to 2018.

Hybrid deep learning and empirical mode decomposition model for time series applications

Time series forecasting is important in many aspects of our lives, since it can be used to deal with the uncertainty to further support the decision making. Despite many advanced methodologies have been proposed, modelling the underlying relationship between past and future conditions is still a challenge. In this research, we develop a novel time series forecasting model which can effectively predict future conditions in a timely fashion. A time series has nonlinear and nonstationary characteristics which make prediction using statistical or computational intelligent methods a difficult task. Therefore, a hybrid deep learning and empirical mode decomposition model, namely EMD–SAE, for multistep ahead forecasting is proposed to predict the traffic flow and random time series. The performances of the proposed model are compared and discussed. This paper shows the potential of hybridizing the deep learning and empirical mode decomposition to the ordinary time series forecasting approach, and the experimental results suggest that the proposed EMD–SAE is reliable, suitable and a promising method for time series forecasting.

Assessment of Time-Series Machine Learning Methods for Forecasting Hospital Discharge Volume

Forecasting the volume of hospital discharges has important implications for resource allocation and represents an opportunity to improve patient safety at periods of elevated risk. To determine the performance of a new time-series machine learning method for forecasting hospital discharge volume compared with simpler methods. A retrospective cohort study of daily hospital discharge volumes at 2 large, New England academic medical centers between January 1, 2005, and December 31, 2014 (hospital 1), or January 1, 2005, and December 31, 2010 (hospital 2), comparing time-series forecasting methods for prediction was performed. Data analysis was conducted from February 28, 2017, to August 30, 2018. Group-level data for all discharges from inpatient units were included. In addition to conventional methods, a technique originally developed for allocating data center resources, and comparison strategies for incorporating prior data and frequency of model updates, was conducted to identify the model application that optimized forecast accuracy. Model calibration as measured by R2 and, secondarily, number of days with errors greater than 1 SD of daily volume. During the forecasted year, hospital 1 had 54 411 discharges (daily mean, 149) and hospital 2 had 47 456 discharges (daily mean, 130). The machine learning method was well calibrated at both sites (R2, 0.843 and 0.726, respectively) and made errors greater than 1 SD of daily volume on only 13 and 22 days, respectively, of the forecast year at the 2 sites. Last-value-carried-forward models performed somewhat less well (calibration R2, 0.781 and 0.596, respectively) with 13 and 46 errors of 1 SD or greater, respectively. More frequent retraining and training sets of longer than 1 year had minimal effects on the machine learning method's performance. Volume of hospital discharges can perhaps be reliably forecasted using simple carry-forward models as well as methods drawn from machine learning. The benefit of the latter does not appear to be dependent on extensive training data and may enable forecasts up to 1 year in advance with superior absolute accuracy to carry-forward models.

Forecasting the orbital-climate diagram dynamics based on the wavelet analysis and fuzzy neural networks

Based on the new concept of the V. Bolshakov's orbital theory of paleoclimate, as well as the multiscale time-series wavelet decomposition method, and neural network fuzzy inference rules, the paper derives a predicted curve for the so-called orbital-climate diagram (OCD) in the ratio of (eccentricity, orbit inclination, precession) within the time interval from -1000 kyr in the past to 100 kyr in the future since modern times. This diagram features the Earth climate change caused by an insolation change to be the principal factor of the climate change, driven by Earth's orbital elements changes. Efficiency of the time-series forecast method is proved by the obtained predicted OCD trajectory verification within the past 100 kyr period and other paleoclimate data with a correlation coefficient 0.93.

Dynamic Similar Sub-Series Selection Method for Time Series Forecasting

Accumulation of influencing factors during several consecutive time periods makes the variation of target parameters lag behind the variation of their influencing factors. This important phenomenon, known as the cumulative effect, would lead to relatively large forecasting errors. In this paper, the dynamic similar sub-series method is proposed to take cumulative effect into consideration. The similar sub-series for the forecasting parameter sub-series are selected based on similarities of target parameter sub-series and influencing factors sub-series. The internal variations of target parameter sub-series and each influencing factor sub-series are innovatively integrated into the selection rules for the dynamic similar sub-series. The corresponding forecasting algorithm is designed, and the forecasting parameters are deduced and forecasted according to the variation of the dynamic similar sub-series. The proposed method is compared with a variety of representative methods under the short-term daily average load forecasting, the electricity price forecasting and the global horizontal irradiance forecasting scenarios to demonstrate its effectiveness.