Search Results (23)

High-variant wind speeds cause aberrations in wind power systems and compromise the effective operation of wind farms. A single model cannot capture the inherent wind speed randomness and complexity. In the proposed hybrid strategy, wavelet transform (WT) is used for data decomposition, sample entropy (SampEn) for subseries complexity evaluation, neural network autoregression (NNAR) for deterministic subseries prediction, long short-term memory network (LSTM) for complex subseries prediction, and gradient boosting machine (GBM) for prediction reconciliation. The proposed WT-NNAR-LSTM-GBM approach predicts minutely averaged wind speed data collected at Southern African Universities Radiometric Network (SAURAN) stations: Council for Scientific and Industrial Research (CSIR), Richtersveld (RVD), Venda, and the Namibian University of Science and Technology (NUST). For comparison purposes, in WT-NNAR-LSTM-GBM, LSTM and NNAR are respectively replaced with a k-nearest neighbour (KNN) to form the corresponding hybrids: WT-NNAR-KNN-GBM and WT-KNN-LSTM-GBM. We assessed WT-NNAR-LSTM-GBM’s efficacy against NNAR, LSTM, WT-NNAR-KNN-GBM, and WT-KNN-LSTM-GBM as well as the naïve model. The comparative study found that the WT-NNAR-LSTM-GBM model was the most accurate, sharpest, and robust based on mean absolute error, median absolute deviation, and residual analysis. The study results suggest using short-term forecasts to optimise wind power production, enhance grid operations in real-time, and open the door to further algorithmic enhancements. Full article

Air pollution, especially ground-level ozone, poses severe threats to human health and ecosystems. Accurate forecasting of ozone concentrations is essential for reducing its adverse effects. This study aims to use the functional time series approach to model ozone concentrations, a method less explored in the literature, and compare it with traditional time series and machine learning models. To this end, the ozone concentration hourly time series is first filtered for yearly seasonality using smoothing splines that lead us to the stochastic (residual) component. The stochastic component is modeled and forecast using a functional autoregressive model (FAR), where each daily ozone concentration profile is considered a single functional datum. For comparison purposes, different traditional and machine learning techniques, such as autoregressive integrated moving average (ARIMA), vector autoregressive (VAR), neural network autoregressive (NNAR), random forest (RF), and support vector machine (SVM), are also used to model and forecast the stochastic component. Once the forecast from the yearly seasonality component and stochastic component are obtained, both are added to obtain the final forecast. For empirical investigation, data consisting of hourly ozone measurements from Los Angeles from 2013 to 2017 are used, and one-day-ahead out-of-sample forecasts are obtained for a complete year. Based on the evaluation metrics, such as $R^2$, root mean squared error (RMSE), and mean absolute error (MAE), the forecasting results indicate that the FAR outperforms the competitors in most scenarios, with the SVM model performing the least favorably across all cases. Full article

The agriculture sector plays an essential function within the Indian economic system. Foodgrains provide almost all the calories and proteins. This paper aims to compare ARIMA, SutteARIMA, Holt-Winters, and NNAR models to recommend an effective model to predict foodgrains production in India. The execution of the SutteARIMA predictive model used in this analysis was compared with the established ARIMA, Neural Network Auto-Regressive (NNAR), and Holt-Winters models, which have been widely applied for time series prediction. The findings of this study reveal that both the SutteARIMA model and the Holt-Winters model performed well with real-life problems and can effectively and profitably be engaged for food grain forecasting in India. The food grain forecasting approach with the SutteARIMA model indicated superior performance over the ARIMA, Holt-Winters, and NNAR models. Indeed, the actual and predicted values of the SutteARIMA and Holt-Winters forecasting models are quite close to predicting foodgrains production in India. This has been verified by MAPE and MSE values that are relatively low with the SutteARIMA model. Therefore, India’s SutteARIMA model was used to predict foodgrains production from 2021 to 2025. The forecasted amount of respective crops are as follows (in lakh tonnes) 1140.14 (wheat), 1232.27 (rice), 466.46 (coarse), 259.95 (pulses), and a total 3069.80 (foodgrains) by 2025. Full article

LSD is an important transboundary disease affecting the cattle industry worldwide. The objectives of this study were to determine trends and significant change points, and to forecast the number of LSD outbreak reports in Africa, Europe, and Asia. LSD outbreak report data (January 2005 to January 2022) from the World Organization for Animal Health were analyzed. We determined statistically significant change points in the data using binary segmentation, and forecast the number of LSD reports using auto-regressive moving average (ARIMA) and neural network auto-regressive (NNAR) models. Four significant change points were identified for each continent. The year between the third and fourth change points (2016–2019) in the African data was the period with the highest mean of number of LSD reports. All change points of LSD outbreaks in Europe corresponded with massive outbreaks during 2015–2017. Asia had the highest number of LSD reports in 2019 after the third detected change point in 2018. For the next three years (2022–2024), both ARIMA and NNAR forecast a rise in the number of LSD reports in Africa and a steady number in Europe. However, ARIMA predicts a stable number of outbreaks in Asia, whereas NNAR predicts an increase in 2023–2024. This study provides information that contributes to a better understanding of the epidemiology of LSD. Full article

During the COVID-19 pandemic, cryptocurrency prices showed abnormal volatility that attracted the participation of many investors. Studying the behaviour of volatility for the prices of cryptocurrency is an interesting problem to be investigated. This research implements the state space model framework for volatility incorporating the Kalman filter. This method directly forecasts the conditional volatility of five cryptocurrency prices (Bitcoin (BTC), Ethereum (ETH), Ripple (XRP), Litecoin (LTC) and Bitcoin Cash (BCH)) for 10,000 consecutive hours, i.e., approximately 417 days during the COVID-19 pandemic from 26 February 2020, 00:00 h until 18 April 2021, 00:00 h. The performance of this model is compared to the GARCH (1,1) model and the neural network autoregressive (NNAR) based on root mean square error (RMSE), mean absolute error (MAE) and the volatility plot. The autocorrelation function plot, histogram and the residuals plot are used to examine the model adequacy. Among the three models, the state space model gives the best fit. The state space model gives the narrowest confidence interval of volatility and value-at-risk forecasts among the three models. Full article

In this paper, a study was completed investigating the financial viability of a 5 MW solar power plant in Montenegro with direct access to the market, rather than a long-term power purchase agreement. The empirical research included an econometric analysis and forecast of the prices on the exchange market, using two methods, autoregressive integrated moving average (ARIMA) and neural network auto regression (NNAR), which are compared to the forecast electricity prices. The former was used in order to obtain the electricity prices forecast, since it showed significantly better predictive performances. Consequently, the financial analysis results indicated this business strategy is a financially more viable option, even though it implies increased risks. All investigated metrics and sensitivity analysis pointed in favor of this option, which has significantly higher profitability with a shorter payback period, compared to the usual market strategy. The main conclusion and recommendation drawn from the analysis are that taking into account the entire environment and prospects for the following years, a riskier business strategy of entering the market directly, or a so-called structured PPA, is put forward to improve project returns and speed up energy-transformation processes in a developing country. Full article

In general, studies on short-term hourly electricity load modeling and forecasting do not investigate in detail the sources of uncertainty in forecasting. This study aims to evaluate the impact and benefits of applying bootstrap aggregation in overcoming the uncertainty in time series forecasting, thereby increasing the accuracy of multistep ahead point forecasts. We implemented the existing and proposed clustering-based bootstrapping methods to generate new electricity load time series. In the proposed method, we use singular spectrum analysis to decompose the series between signal and noise to reduce the variance of the bootstrapped series. The noise is then bootstrapped by K-means clustering-based generation of Gaussian normal distribution (KM.N) before adding it back to the signal, resulting in the bootstrapped series. We apply the benchmark models for electricity load forecasting, SARIMA, NNAR, TBATS, and DSHW, to model all new bootstrapped series and determine the multistep ahead point forecasts. The forecast values obtained from the original series are compared with the mean and median across all forecasts calculated from the bootstrapped series using the Malaysian, Polish, and Indonesian hourly load series for 12, 24, and 36 steps ahead. We conclude that, in this case, the proposed bootstrapping method improves the accuracy of multistep-ahead forecast values, especially when considering the SARIMA and NNAR models. Full article

Thailand is one of the countries where foot and mouth disease outbreaks have resulted in considerable economic losses. Forecasting is an important warning technique that can allow authorities to establish an FMD surveillance and control program. This study aimed to model and forecast the monthly number of FMD outbreak episodes (n-FMD episodes) in Thailand using the time-series methods, including seasonal autoregressive integrated moving average (SARIMA), error trend seasonality (ETS), neural network autoregression (NNAR), and Trigonometric Exponential smoothing state–space model with Box–Cox transformation, ARMA errors, Trend and Seasonal components (TBATS), and hybrid methods. These methods were applied to monthly n-FMD episodes (n = 1209) from January 2010 to December 2020. Results showed that the n-FMD episodes had a stable trend from 2010 to 2020, but they appeared to increase from 2014 to 2020. The outbreak episodes followed a seasonal pattern, with a predominant peak occurring from September to November annually. The single-technique methods yielded the best-fitting time-series models, including $SARIMA\left(1,0,1\right){\left(0,1,1\right)}_{12}$, $NNAR{\left(3,1,2\right)}_{12}$,$ETS\left(A,N,A\right)$, and $TBATS(1,\left\{0,0\right\},0.8,\left\{<12,5>\right\}$. Moreover, SARIMA-NNAR and NNAR-TBATS were the hybrid models that performed the best on the validation datasets. The models that incorporate seasonality and a non-linear trend performed better than others. The forecasts highlighted the rising trend of n-FMD episodes in Thailand, which shares borders with several FMD endemic countries in which cross-border trading of cattle is found common. Thus, control strategies and effective measures to prevent FMD outbreaks should be strengthened not only in Thailand but also in neighboring countries. Full article

In December 2021, the latest COVID-19 variant, Omicron, was confirmed in Indonesia. Unlike the Delta variant, the number of deaths in the Omicron type did not increase significantly and remained constant, even though the cases increased significantly. It is hoped that Indonesia will declare COVID-19 endemic. Jakarta is the capital of Indonesia and the first city where the new COVID-19 virus emerged. Therefore, we are trying to model COVID-19 cases in Jakarta and predict future cases to see if endemic conditions are identified. We applied Neural Network Auto-Regressive (NNAR) and Long Short-Term Memory (LSTM) methods. It is found that the NNAR forecast better for positive confirmed cases with an R-squared 0.939 and the LSTM forecast better for cases of death with an R-squared 0.9337. The forecasting results for the next 7 days reveal that positive confirmed cases of COVID-19 in Jakarta will increase slightly. In addition, the death cases experienced a very small increase, only one new case. According to the results of this study, it can be concluded that COVID-19 in Jakarta will enter an endemic phase in Jakarta, with no substantial increase in cases and a low mortality rate. Full article

Cancer remains a leading cause of worldwide mortality and is a growing, multifaceted global burden. As a result, cancer prevention and cancer mortality reduction are counted among the most pressing public health issues of the twenty-first century. In turn, accurate projections of cancer incidence and mortality rates are paramount for robust policymaking, aimed at creating efficient and inclusive public health systems and also for establishing a baseline to assess the impact of newly introduced public health measures. Within the European Union (EU), Romania consistently reports higher mortality from all types of cancer than the EU average, caused by an inefficient and underfinanced public health system and lower economic development that in turn have created the phenomenon of “oncotourism”. This paper aims to develop novel cancer incidence/cancer mortality models based on historical links between incidence and mortality occurrence as reflected in official statistics and population web-search habits. Subsequently, it employs estimates of the web query index to produce forecasts of cancer incidence and mortality rates in Romania. Various statistical and machine-learning models—the autoregressive integrated moving average model (ARIMA), the Exponential Smoothing State Space Model with Box-Cox Transformation, ARMA Errors, Trend, and Seasonal Components (TBATS), and a feed-forward neural network nonlinear autoregression model, or NNAR—are estimated through automated algorithms to assess in-sample fit and out-of-sample forecasting accuracy for web-query volume data. Forecasts are produced with the overperforming model in the out-of-sample context (i.e., NNAR) and fed into the novel incidence/mortality models. Results indicate a continuation of the increasing trends in cancer incidence and mortality in Romania by 2026, with projected levels for the age-standardized total cancer incidence of 313.8 and the age-standardized mortality rate of 233.8 representing an increase of 2%, and, respectively, 3% relative to the 2019 levels. Research findings thus indicate that, under the no-change hypothesis, cancer will remain a significant burden in Romania and highlight the need and urgency to improve the status quo in the Romanian public health system. Full article

Predicting and modeling time-to-events data is a crucial and interesting research area. For modeling and predicting such types of data, numerous statistical models have been suggested and implemented. This study introduces a new statistical model, namely, a new modified flexible Weibull extension (NMFWE) distribution for modeling the mortality rate of COVID-19 patients. The introduced model is obtained by modifying the flexible Weibull extension model. The maximum likelihood estimators of the NMFWE model are obtained. The evaluation of the estimators of the NMFWE model is assessed in a simulation study. The flexibility and applicability of the NMFWE model are established by taking two datasets representing the mortality rates of COVID-19-infected persons in Mexico and Canada. For predictive modeling, we consider two pure statistical models and two machine learning (ML) algorithms. The pure statistical models include the autoregressive moving average (ARMA) and non-parametric autoregressive moving average (NP-ARMA), and the ML algorithms include neural network autoregression (NNAR) and support vector regression (SVR). To evaluate their forecasting performance, three standard measures of accuracy, namely, root mean square error (RMSE), mean absolute error (MAE), and mean absolute percentage error (MAPE) are calculated. The findings demonstrate that ML algorithms are very effective at predicting the mortality rate data. Full article

The European Union (EU) has positioned itself as a frontrunner in the worldwide battle against climate change and has set increasingly ambitious pollution mitigation targets for its members. The burden is heavier for the more vulnerable economies in Central and Eastern Europe (CEE), who must juggle meeting strict greenhouse gas emission (GHG) reduction goals, significant fossil-fuel reliance, and pressure to respond to current pandemic concerns that require an increasing share of limited public resources, while facing severe repercussions for non-compliance. Thus, the main goals of this research are: (i) to generate reliable aggregate GHG projections for CEE countries; (ii) to assess whether these economies are on track to meet their binding pollution reduction targets; (iii) to pin-point countries where more in-depth analysis using spatial inventories of GHGs at a finer resolution is further needed to uncover specific areas that should be targeted by additional measures; and (iv) to perform geo-spatial analysis for the most at-risk country, Poland. Seven statistical and machine-learning models are fitted through automated forecasting algorithms to predict the aggregate GHGs in nine CEE countries for the 2019–2050 horizon. Estimations show that CEE countries (except Romania and Bulgaria) will not meet the set pollution reduction targets for 2030 and will unanimously miss the 2050 carbon neutrality target without resorting to carbon credits or offsets. Austria and Slovenia are the least likely to meet the 2030 emissions reduction targets, whereas Poland (in absolute terms) and Slovenia (in relative terms) are the farthest from meeting the EU’s 2050 net-zero policy targets. The findings thus stress the need for additional measures that go beyond the status quo, particularly in Poland, Austria, and Slovenia. Geospatial analysis for Poland uncovers that Krakow is the city where pollution is the most concentrated with several air pollutants surpassing EU standards. Short-term projections of PM2.5 levels indicate that the air quality in Krakow will remain below EU and WHO standards, highlighting the urgency of policy interventions. Further geospatial data analysis can provide valuable insights into other geo-locations that require the most additional efforts, thereby, assisting in the achievement of EU climate goals with targeted measures and minimum socio-economic costs. The study concludes that statistical and geo-spatial data, and consequently research based on these data, complement and enhance each other. An integrated framework would consequently support sustainable development through bettering policy and decision-making processes. Full article

Forecasting the industry’s electricity consumption is essential for energy planning in a given country or region. Thus, this study aims to apply time-series forecasting models (statistical approach and artificial neural network approach) to the industrial electricity consumption in the Brazilian system. For the statistical approach, the Holt–Winters, SARIMA, Dynamic Linear Model, and TBATS (Trigonometric Box–Cox transform, ARMA errors, Trend, and Seasonal components) models were considered. For the approach of artificial neural networks, the NNAR (neural network autoregression) and MLP (multilayer perceptron) models were considered. The results indicate that the MLP model was the one that obtained the best forecasting performance for the electricity consumption of the Brazilian industry under analysis. Full article

Climate change and pollution fighting have become prominent global concerns in the twenty-first century. In this context, accurate estimates for polluting emissions and their evolution are critical for robust policy-making processes and ultimately for solving stringent global climate challenges. As such, the primary objective of this study is to produce more accurate forecasts of greenhouse gas (GHG) emissions. This in turn contributes to the timely evaluation of the progress achieved towards meeting global climate goals set by international agendas and also acts as an early-warning system. We forecast the evolution of GHG emissions in 12 top polluting economies by using data for the 1970–2018 period and employing six econometric and machine-learning models (the exponential smoothing state-space model (ETS), the Holt–Winters model (HW), the TBATS model, the ARIMA model, the structural time series model (STS), and the neural network autoregression model (NNAR)), along with a naive model. A battery of robustness checks is performed. Results confirm a priori expectations and consistently indicate that the neural network autoregression model (NNAR) presents the best out-of-sample forecasting performance for GHG emissions at different forecasting horizons by reporting the lowest average RMSE (root mean square error) and MASE (mean absolute scaled error) within the array of predictive models. Predictions made by the NNAR model for the year 2030 indicate that total GHG emissions are projected to increase by 3.67% on average among the world’s 12 most polluting countries until 2030. Only four top polluters will record decreases in total GHG emissions values in the coming decades (i.e., Canada, the Russian Federation, the US, and China), although their emission levels will remain in the upper decile. Emission increases in a handful of developing economies will see significant growth rates (a 22.75% increase in GHG total emissions in Brazil, a 15.75% increase in Indonesia, and 7.45% in India) that are expected to offset the modest decreases in GHG emissions projected for the four countries. Our findings, therefore, suggest that the world’s top polluters cannot meet assumed pollution reduction targets in the form of NDCs under the Paris agreement. Results thus highlight the necessity for more impactful policies and measures to bring the set targets within reach. Full article

The present study employs daily data made available by the STR SHARE Center covering the period from 1 January 2010 to 31 January 2020 for six Viennese hotel classes and their total. The forecast variable of interest is hotel room demand. As forecast models, (1) Seasonal Naïve, (2) Error Trend Seasonal (ETS), (3) Seasonal Autoregressive Integrated Moving Average (SARIMA), (4) Trigonometric Seasonality, Box–Cox Transformation, ARMA Errors, Trend and Seasonal Components (TBATS), (5) Seasonal Neural Network Autoregression (Seasonal NNAR), and (6) Seasonal NNAR with an external regressor (seasonal naïve forecast of the inflation-adjusted ADR) are employed. Forecast evaluation is carried out for forecast horizons h = 1, 7, 30, and 90 days ahead based on rolling windows. After conducting forecast encompassing tests, (a) mean, (b) median, (c) regression-based weights, (d) Bates–Granger weights, and (e) Bates–Granger ranks are used as forecast combination techniques. In the relative majority of cases (i.e., in 13 of 28), combined forecasts based on Bates–Granger weights and on Bates–Granger ranks provide the highest level of forecast accuracy in terms of typical measures. Finally, the employed methodology represents a fully replicable toolkit for practitioners in terms of both forecast models and forecast combination techniques. Full article