Forecasting Time Series Data Research Articles

Deep learning and optimisation for quality of service modelling

Machine learning is increasingly used to create digital twins for data collected from various underlying engineering processes. Such digital twins can be used in a wide variety of activities such as optimisation, forecasting of future data, etc. In this respect, forecasting the evolution of time-series data in the future time-steps is often encountered in various engineering systems and applications. In particular, probabilistic forecasting of time-series data over point-based predictions is often encouraged, but challenging to achieve though. In this work, deep learning (DR) technology is combined with various state-of-the-art mathematical optimisation algorithms in order to effectively achieve the ’confidence-based’ probabilistic predictions of Quality of Service (QoS) data emanating from various low-powered Internet of Things (IoT) devices. The results demonstrate that Deep Neural Networks (DNN), if combined with right mathematical optimisation algorithm, can help generating accurate probabilistic forecasts for both single time-series and a combination of multiple time-series data.

Analysis of Rice Production Forecast in Maros District Using the Box-Jenkins Method with the ARIMA Model

The Box-Jenkins method is a statistical method used for forecasting time series data. This method uses data in the past as the dependent variable. The data used in this study is data on the amount of rice production in Maros Regency which was taken from 2001 to 2018 which was taken from the Central Statistics Agency of Maros Regency and the Department of Food Security, Food Crops, and Horticulture of South Sulawesi Province. The results obtained show that the ARIMA(0,2,1) model is a suitable model to predict the amount of rice production in Maros Regency. Forecasting results show that the amount of rice production in Maros Regency has increased every year with an average increase of 3807.1 tons.

Non-linear autoregressive neural network with exogenous variable in forecasting USD/IDR exchange rate

In the history of economic development, the Rupiah exchange rate is considered one of the key factors in economic stability in Indonesia. Bank Indonesia’ policies play an important role in the Indonesia's economy. The purpose of this study is to explore the relationship between time effect and the presence of exogenous variable that affect the response variable in the non-linear model. The methodological approach taken in this study is Nonlinear Autoregressive Exogenous Neural Network Method (NARX NN). NARX NN is a powerful method for forecasting of time series data and dynamic control problems. NARX NN method extracts generic principles from the past values of the time series to predict its future values so that can be used to forecast the Rupiah exchange rate to the dollar based on the BI rate variable. The obvious finding to emerge from this study are: 1) Feed Forward Neural Network (FFNN) with rprop+ training algorithm, the best model is FFNN (12-3-1). The best FFNN model applied to all BI 7-day (Reverse) Repo Rate data has an Adjusted R 2 of 98.8%; 2) Forecasting the USD/IDR exchange rate and its relation to the BI 7-day (Reverse) Repo Rate using the NARX NN series parallel model with the rptop+ training algorithm, obtained the best model NARX NN (13-4-1). The NARX NN model applied to all USD/IDR exchange rate data has an Adjusted R 2 of 96.19%; 3) Based on forecasting results for the next 6 periods, the USD/IDR exchange rate tends to experience a downward trend, meaning that the Rupiah exchange rate strengthens, while the BI 7-day (Reverse) Repo Rate increases.

A Comparative Simulation Study of ARIMA and Computational Intelligent Techniques for Forecasting Time Series Data

A Comparative Simulation Study of ARIMA and Computational Intelligent Techniques for Forecasting Time Series Data

Data analytics and optimised machine learning algorithm to analyse coffee commodity prices

Coffee has long become a familiar and famous drink all over the world. Due to supply chain disruptions, the coffee market is vibrant and changing rapidly. In this article, the time series data on world coffee price will be analysed by using data analytics such as Lyapunov exponent (LE), entropy, and Hurst exponent (HE). By denoising time series data using wavelet decomposition, the echo state network (ESN) model is presented for forecasting time series data. When the coffee commodity price is fluctuating and affecting to production planning and scheduling of supply chain management under market uncertainty, it is necessary to improve the accuracy and efficiency of the prediction method by employing the grey wolf optimisation (GWO) algorithm. By employing the hybrid technique, the predictive analytics models could efficiently estimate the actual prices. The GWO algorithm enhances a machining learning model's performance for ensuring optimal forecasting with increased accuracy.

Study of Various Forecasting Models for time series data, using Stochastic Processes

The data which is in time stamped format is called as time series data. The time series data is everywhere for example Weather data, Stock market data, health care data, Sensor data, network data, sales data and many more. Time series have various components due to which the time series data became complex. Trend, Seasonality, Cyclical, and irregularities, these are different components. As everyone interested to know about future. That’s why Forecasting using time series data is important point of consideration. This research paper focuses on components of time series data simultaneously study of different time series modelling and forecasting techniques which are based on stochastic processes. Mainly all the models discussed here focus on use of past time series data for forecasting future values. The Research paper covers AR, MA, Random Walk, ARMA, ARIMA, SARIMA, and Exponential Smoothing processes (single, double and triple) which are used for forecasting time series data.

Analysis and Application of Grey-Markov Chain Model in Tax Forecasting

Tax data is a typical time series data, which is subject to the interaction and influence of economic and political factors and has dynamic and highly nonlinear characteristics. The key to correct tax forecasting is the choice of forecasting algorithm. Traditional tax forecasting methods, such as factor scoring method, factor regression method, and system adjustment method, have a certain guiding role in actual work, but there are still many shortcomings, such as the limitation from the distribution and size of sample data and difficulty of grasping the nonlinear phenomena in economic system. Grey-Markov chain model formed by the combination of grey forecasting and Markov chain forecasting can not only reveal the general developmental trend of time series data, but also predict their state change patterns. Based on the summary and analysis of previous research works, this paper expounds the current research status and significance of tax forecasting, elaborates the development background, current status, and future challenges of the Grey-Markov chain model, introduces the basic principles of grey forecasting model and Markov chain model, constructs the Grey-Markov chain model, analyzes the model’s residual error and posteriori error tests, conducts the analysis of Grey-Markov chain model, performs grey forecasting model construction and its state division, implements the calculation of transition probability matrix and the determination of tax forecasting value, discusses the application of the Grey-Markov chain model in tax forecasting, and finally carries out a simulation experiment and its result analysis. The study results show that, compared with separate grey forecasting, Markov chain forecasting, and other commonly used time series forecasting methods, the Grey-Markov chain model increases the accuracy of tax forecasts by an average of 2.3–13.1%. This indicates that the combinative forecasting of Grey-Markov chain model can make full use of the information provided by time series data for tax analysis and forecasting. It can not only avoid the influence of economic, political, and human subjective factors, but also have simple calculations, higher accuracy, and stronger practicality. The study results of this paper provide a reference for further researches on the analysis and application of Grey-Markov chain model in tax forecasting.

PERBANDINGAN METODE SEASONAL ARIMA DAN EXTREME LEARNING MACHINE PADA PERAMALAN JUMLAH WISATAWAN MANCANEGARA KE BALI

Bali Island of the Gods is one of the wealth of very popular tourist destinations and has the highest number of foreign tourists in Indonesia. It is very necessary to do more in-depth learning related to the projections or forecasting of foreign tourist visits to Bali at a certain period of time. Forecasting analysis used is to compare two methods, namely the Seasonal ARIMA method (SARIMA) and Extreme Learning Machine (ELM). The SARIMA method is a statistical method commonly used in forecasting time series data that contains seasonality and has good accuracy. While the ELM method is a new learning method of artificial neural networks that has fast learning speed and good accuracy. The results obtained indicate that the Seasonal ARIMA method is a better method used to predict the number of tourists to Bali in this case, because it has a smaller forecasting MAPE value of 4.97%. While the ELM method has a forecasting MAPE value of 7.62%.

PENDUGAAN PARAMETER MODEl DISTRIBUTED LAG POLA POLINOMIAL MENGGUNAKAN METODE ALMON

The distributed lag model is a regression model that describes the relationship between the dependent variable of a given period and the independent variables of a certain or previous periods. The model can be used to determine the impact of the independent variable to dependent variables over time and forecast time series data for the next periods. There are two forms of distributed lag model that have been widely proposed in the estimation of distributed lag regression model. The first form is proposed by Koyck and the second form by Almon. This paper aims to apply the Almon model to examine the effect of the ratio of BOPO (Operating Cost and Operating Income) to the ROA (Return on Asset) of a government bank based on quarterly data, to estimate its parameters, to examine the feasibility of the model, and to predict the next quarter. Results shows that distributed lag model is = 10.110 - 0.078 + 0.015 + 0.026 – 0.045 with Yt is ROA, and Xt is the ratio BOPO on the 1st quarter until the previous 3 quarters. The model is quite good according to the determination coefficient that is 0.75, no autocorrelation in the model, t test and F test are also significant. Based on the model, the value of ROA ratio next quarter predicted 4.63%. The decrease in profitability ROA ratio is due to an increase in interest expense while interest income can not compensate

Modeling and Forecasting of Time Series Data using Different Techniques

Modeling and Forecasting of Time Series Data using Different Techniques

Time series prediction using machine learning: a case of Bitcoin returns

PurposeThe purpose of this study is to compare five data-driven-based ML techniques to predict the time series data of Bitcoin returns, namely, alternating model tree, random forest (RF), multiple linear regression, multi-layer perceptron regression and M5 Tree algorithms.Design/methodology/approachThe data used to forecast time series data of Bitcoin returns ranges from 8 July 2010 to 30 Aug 2020. This study used several predictors to predict bitcoin returns including economic policy uncertainty, equity market volatility index, S&P returns, USD/EURO exchange rates, oil and gold prices, volatilities and returns. Five statistical indexes, namely, correlation coefficient, mean absolute error, root mean square error, relative absolute error and root relative squared error are determined. The results of these metrices are used to develop colour intensity ranking.FindingsAmong the machine learning (ML) techniques used in this study, RF models has shown superior predictive ability for estimating the Bitcoin returns.Originality/valueThis study is first of its kind to use and compare ML models in the prediction of Bitcoins. More studies can be carried out by using further cryptocurrencies and other ML data-driven models in future.

PAN-LDA: A latent Dirichlet allocation based novel feature extraction model for COVID-19 data using machine learning

The recent outbreak of novel Coronavirus disease or COVID-19 is declared a pandemic by the World Health Organization (WHO). The availability of social media platforms has played a vital role in providing and obtaining information about any ongoing event. However, consuming a vast amount of online textual data to predict an event's trends can be troublesome. To our knowledge, no study analyzes the online news articles and the disease data about coronavirus disease. Therefore, we propose an LDA-based topic model, called PAN-LDA (Pandemic-Latent Dirichlet allocation), that incorporates the COVID-19 cases data and news articles into common LDA to obtain a new set of features. The generated features are introduced as additional features to Machine learning(ML) algorithms to improve the forecasting of time series data. Furthermore, we are employing collapsed Gibbs sampling (CGS) as the underlying technique for parameter inference. The results from experiments suggest that the obtained features from PAN-LDA generate more identifiable topics and empirically add value to the outcome.

Predictability of second-hand bulk carriers with a novel hybrid algorithm

This paper investigates the predictability of the asset prices of commodity transport (i.e. dry bulk carriers) by testing the shipping Q index as a leading indicator. We employ a comprehensive back-testing procedure with a broad spectrum of benchmark simulations. The shipping Q index (an adaptation of Tobin's Q index) has been introduced to benchmark models to observe predictive gain and interpret predictability features. This study presents a novel hybrid model to forecast time series data. The forecasting ability of the proposed hybrid algorithm is compared to specific univariate time series models, dynamic models, nonlinear models, and widely used hybrid models in the literature. The findings document that not only the proposed hybrid model performs better than the other competitive models in terms of hold out sample forecasting, but also using the shipping Q index improves the forecast accuracy by remarkably reducing forecasting error.

Study and analysis of SARIMA and LSTM in forecasting time series data

Energy consumption forecasting is essential for smart grid operations as it facilitates electricity demand management and utilities load planning. In this paper data analytics has been presented on the collected smart meter measurement and then predicting the energy consumption on a daily basis using (autoregressive integrated moving average) ARIMA, seasonal ARIMA (SARIMA) and long short-term memory (LSTM). The analysis tends to understand the different factors which influence energy consumption, and assist operators to make decisions accordingly. It is helpful in reducing the outage, and enhancing the situational awareness of power consumption on a daily basis of the smart meters. The relational factors are capable in lowering energy consumption, or rather contributing to the effective consumption of energy units. The parameters used for the result evaluation are various features of the weather features relation in terms of power consumption based on temperature, humidity, cloud cover, visibility, wind speed, UV index and dew point. The results indicate that the energy consumption has a high positive correlation with humidity and high negative correlation with temperature. (Dew point and UV index) and (Cloud cover and Visibility Display) have multicollinearity with temperature and humidity respectively, so, can be discarded. Pressure and Moon Phase have minimal correlation with energy consumption, so, it can also be discarded. Wind speed has low correlation with energy, but it does not show multicollinearity. So, it can be considered for further analysis. Overall LSTM found to be prominent in comparison to ARIMA and SARIMA with the average mean absolute error (MAE) of 0.23.

Time series modeling of road traffic accidents in Amhara Region

Road traffic accidents (RTA) are commonly encountered incidents that can cause injuries, death, and property damage to members of society. Ethiopia is one of the highest incident rates of road traffic accidents. Report of Transport and Communication from 2012 to 2014, shows an increment in the number of traffic accidents in Ethiopia. Amhara region accounted for 27.3% of the total road traffic accident-related deaths in Ethiopia during the year 2008/9, which is the highest share among all regions in Ethiopia. The current research aims to model the trend of injury, fatal and total road traffic accidents in the Amhara region from September 2013 to May 2017. Monthly reported traffic accidents were obtained from the traffic department of the Amhara region police commission. The most universal class of models for forecasting time series data called Auto-regressive Integrated Moving Averages (ARIMA) models were applied to model the trends and patterns of road traffic accident cases in the Amhara region. The average number of observed injury RTA, fatal RTA, and total RTA were 27.2, 14, and 78.2 per month respectively. It was observed that a relatively large number of RTA’s are reported on Tuesday, Thursday, and Saturday relative to other days of the week. The data also reveals that more than 60% of accidents involve drivers between the ages of 18–30 years. ARIMA (2,0,0) (1,0,0) ARIMA (2,0,0) and ARIMA (2,0,0) (1,1,0) were fitted as the best model for total injury accidents, fatal RTA and total RTA data respectively. A 48 months forecast was made based on the fitted models and it can be concluded that road traffic accident cases would continue at the non-decreasing rate in the Amhara region for the predicted periods. Therefore, the findings of this study draw attention to the importance of implementing improved better policies and close monitoring of road trafficking to change the existing non-decreasing trend of road traffic accidents in the region.

A non-parametric model for fuzzy forecasting time series data

A non-parametric model for fuzzy forecasting time series data

Forecasting time series data containing outliers with the ARIMA additive outlier method

ARIMA method is often used in time series data forecasting. However, when an outlier occurs or there is an observation value that is far from a set of data, forecasting with this method will provide a large residual so that the normality assumption is not met. One method developed from the ARIMA Method that overcomes outliers, is called the ARIMA Additive Outlier Method (ARIMA AO). The purpose of this study is to forecast time series data containing outliers with the ARIMA AO Method. This method can reduce the presence of outliers with iterative procedures. The data used is data on the number of foreign tourists visiting the Port of Tanjung Priok. The results obtained are that the model has a normal distribution residual and forecasting accuracy value with MSE and MAPE is smaller than the ARIMA.

AN INVESTIGATION INTO THE PERFORMANCE OF THE MULTILAYER PERCEPTRON ARCHITECTURE OF DEEP LEARNING IN FORECASTING STOCK PRICES

A wide range of studies have been conducted on deep learning to forecast time series data. However, very few researches have discussed the optimal number of hidden layers and nodes in each hidden layer of the architecture. It is crucial to study the number of hidden layers and nodes in each hidden layer as it controls the performance of the architecture. Apart from that, in the presence of the activation function, diverse computation between the hidden layers and output layer can take place. Therefore, in this study, the multilayer perceptron (MLP) architecture is developed using the Python software to forecast time series data. Then, the developed architecture is applied on the Apple Inc. stock price due to its volatile characteristic. Using historical prices, the accuracy of the forecast is measured by the different activation functions, number of hidden layers and size of data. The Keras deep learning library, which can be found in the Python software, is used to develop the MLP architecture to forecast the Apple Inc. stock price. The developed model is then applied on different cases, namely different sizes of data, different activation functions, different numbers of hidden layers of up to nine layers, and different numbers of nodes in each hidden layer. Then, the metrics mean squared error (MSE), mean absolute error (MAE) and root-mean-square error (RMSE) are employed to test the accuracy of the forecast. It is found that the architecture with rectified linear unit (ReLU) outperformed in every hidden layer and each case with the highest accuracy. To conclude, the optimal number of hidden layers differs in every case as there are other influencing factors.

Regularized Noise based GRU Model to Forecast Solid Waste Generation in the Urban Region

Today’s Solid Waste Management System needs better strategies to perform effective operations. The speed at which waste is being generated in urban regions is directly affecting the surrounding environment impacting the health of living beings. Advancements in deep learning has made time series forecasting very efficient making better predictions. Recurrent Neural Network (RNN) is a deep learning approach which helps in analyzing and forecasting time series data by holding the complication of sequence dependencies. In this study, a RNN model, Regularized Noise based Gated Recurrent Unit (RNGRU) has been proposed to analyze and predict solid waste generation in Australia. The dataset is collected from www. oecd.org which has waste data 20780 records from 1990 to 2018. Various input features such as Region name, Year, Waste generated in tonnes and technical indicators such as Moving Average (MA), Exponential Moving Average (EMA), Moving Average Convergence/Divergence (MACD), Relative Strength Index (RSI), On Balance Volume (OBV), Momentum (MTM), Daily waste variation are used to train the model. The proposed RNGRU model is compared with LSTM model and the best model is chosen based on the performance metrics, Mean Absolute Error(MAE), Mean Squared Error (MSE), and Root Mean Squared Error(RMSE). The outcome of the experiments shows that the RNGRU model is best for prediction with low error rates, MAE value 0.0147, MSE value 0.0010 and RMSE value 0.0900 respectively.

Artificial Neural Network Model with PSO as a Learning Method to Predict Movement of the Rupiah Exchange Rate against the US Dollar

The movement of currency exchange rate can be predicted in the next few days, this is used by economic actors to get profit. Artificial Neural Network with the backpropagation learning method is good enough to use for forecasting time series data, it's just that in its application this method was considered to have shortcomings such as a long training time to achieve convergence. The purpose of this research is to form a Multilayer Perceptron Artificial Neural Network model with the Particle Swarm Optimization (PSO) algorithm as a learning method in the case of currency exchange rate prediction. This research produced a model that can predict the movement of the Rupiah exchange rate against the US Dollar, while the model formed was the MLP-PSO model with an error rate of 5.6168 x 10-8, slightly better than the MLP-BP model with an error rate of 6.4683 x 10-8. These results indicated that the PSO algorithm can be used as a learning algorithm in the Multilayer Perceptron Artificial Neural Network.