جستجوی مقالات مرتبط با کلیدواژه « time series forecasting » در نشریات گروه « فنی و مهندسی »

Financial time series are fundamentally complex, dynamic, noisy, non-linear, non-parametric and chaotic, so forecasting financial markets is one of the most challenging fields in engineering and economics. With the increasing progress of artificial intelligence and the emergence of deep learning methods, the problem of stock market forecasting has been faced with significant developments, especially in the field of prediction models and big data. Four important steps to create a systematic intelligent forecasting model include model inputs, selection of forecasting algorithms and design of the general structure of the forecasting model, using appropriate loss functions to train the algorithm and finally suitable evaluation of the results, according to the desired criteria. In this paper, a comprehensive review of recent approaches to stock market forecasting is provided, focusing on the above steps. The most important achievements of this paper are: 1- A comprehensive review of the problem, including: reviewing the types of model inputs, different prediction structures, training the model and types of loss functions used, and the evaluation metric of the results, in a fully classified and structured way to easily provide the road map and existing challenges for the enthusiasts and also an important research field of each section for the researchers. 2- Complete analysis of each part, specifying the application of each method and discussing their advantages and disadvantages based on the latest developments and providing perspectives on the research boundaries. 3- Determining the ongoing research path, future approaches and open issues for researchers interested in this field.

Coronavirus (COVID-19) is a pandemic that has affected all countries of the world. Forecasting the spread of corona disease will lead to the necessary measures to be taken by the authorities to control this disease. These include increasing vaccinations, quarantining cities and banning entry and exit, increasing the capacity of hospital beds, setting up round-the-clock vaccination centers, requiring the use of masks in public places, and observing social distances. Therefore, predicting such cases will reduce the number of corona cases and therefore reduce the mortality rate.

In this paper, using the Singular Spectrum Analysis (SSA) algorithm, the sixth peak of coronavirus in Iran is predicted by considering the current situation. To improve the grouping process of the SSA algorithm, eigenvalues have been selected in the optimization process, so that the predicted time series of which has been significantly improved according to the error-index.

Comparing the proposed method with other forecasting methods include Autoregressive Integrated Moving Average (ARIMA), Fractional ARIMA (ARFIMA), TBATS, and Neural Network Autoregression (NNAR), it is observed that the forecasting error is acceptable and the SSA method can be used for forecasting.

Originality/Value: 

This article predicts a new case of COVID-19 using efficient method SSA and the presented results confirm the effectiveness of the proposed method.

This paper proposes a remaining useful life (RUL) prediction method that uses the peak of the vibration acceleration signal as an appropriate feature to indicate the degradation process in the rolling element bearings (REBs). In the first step, this feature is transformed into a stationary time series using logarithmic transformation. That is because the long short-term memory neural network (LSTM-NN) works better with the stationary time series. Training the LSTM-NN is performed by this stationary time series as the input and the response is the training time series with values shifted by one time step. Therefore, the LSTM-NN learns to predict the value of the next time step at each point. In other words, to forecast the values of multiple time steps in the future, previous forecasted steps are used as inputs. Next, the values of the future time steps are returned to the main non-stationary form to predict the trend of the peak in the future. Importantly, new measured data can be used to perform new predictions. For this purpose, for every new measured data, the LSTM-NN repeats the mentioned steps and generates a new trend. This algorithm is a trend-dependent method. Therefore, an REB that has a slow degradation stage in its life, which is corresponding to the growth and expansion of defects in REBs, is appropriate to be studied by this algorithm. This method is implemented on two REBs from PRONOSTIA accelerated-life test which have been used by many researchers in the literature. According to the prediction results, the remaining time that peak amplitude trend touches a given threshold is provided. If this threshold is a criterion for the end of life (EoL), this method can be used to determine the RUL. The performance of the proposed method has been evaluated and the presented results are in a good agreement with the experimental data.

Forecasting methods are one of the most efficient available approaches to make managerial decisions in various fields of science. Forecasting is a powerful approach in the planning process, policy choices and economic performance. The accuracy of forecasting is an important factor affects the quality of the decisions that generally has direct and non-strict relationship with the quality of decisions. This is the most important reason that why endeavor for improving the forecasting accuracy has never been stopped in the literature. Electricity demand forecasting is one of the most challenging areas forecasting and important factors in the management of energy systems and economic performance. Determining the level of electricity demand is essential for careful planning and implementation of the necessary policies. For this reason electricity demand forecasting is important for financial and operational managers of electricity distribution. The unique feature of the electricity which makes it more difficult forecasting in comparison with other commodity is the impossibility of storing it in order to use in the future. In other words, the production and consumption of electricity should be taken simultaneously. It has caused to create a high level of complexity and ambiguity in electricity markets data. Computational intelligence and soft computing approaches are among the most precise and useful methods for modeling the complexity and uncertainty in data. In this paper a soft intelligent method by combining mentioned methods is proposed in order to electricity demand forecasting. The main idea of the proposed model is to simultaneously use advantages of these models in modeling complex and ambiguous systems. Empirical results indicate that proposed model can achieve more accurate results rather than its component (Seasonal auto-regressive Integrated Moving Average models, artificial neural network) and also other current single forecasting methods such as classic regression, Seasonal Auto-Regressive Integrated Moving Average-fuzzy models and support vector machine

Decision-making as one of the principles of management is considered an important factor in prosperity of the organizations. This is so important that managers use efficient tools to improve the quality of their decisions. Steel industry is one of the major industries in this country; consequently, it deserves special attention. In this paper, the main aim is to use scientific methods to manage crude steel consumption in the country. However, the literature shows that it is relatively difficult to yield accurate results in the prediction of consumption, especially in long-term horizon. Researchers believe that high level of complexity and uncertainty in financial markets is main reason of this matter. Therefore, in this paper, a hybrid of intelligent and soft computing models have been used as an effective way in order to model the complexities and uncertainties simultaneously in the data. In this way, the list of variables is recognized based on the literature and expert opinions. Then the linear and nonlinear relationships and also correlations between variables are evaluated and final explanatory variables specified. Finally, four models including hard classic, soft classic, hard intelligent and soft intelligent are designed to predict steel consumption in both short and long term horizons and their results are compared with each other. Empirical results indicate that using the hard intelligent model makes improvement 22.68% and 41.41% in comparison with hard classic model in short and long term horizons respectively in Root Mean Squared Error (RMSE). In addition, the soft intelligent model makes improvement 43.01% and 92.72% in comparison with soft classic model and hard classic model respectively in short term horizon and 34.68% and 91.53% in long term horizon. Results of the study indicate superiority of the soft intelligent models over hard intelligent models and superiority of hard intelligent models over hard classic models. Results of the study indicate superiority of the soft intelligent models and hard intelligent models over hard intelligent models and hard classic models respectively.