پیش‎بینی

This study examines the accuracy of Heterogeneous Autoregressive (HAR) models in forecasting the Conditional Value-at-Risk (CVaR) of Exchange-Traded Funds (ETFs) on the Tehran Stock Exchange. The significance of this study stems from the need for better risk management in financial markets, where volatility and jumps significantly affect investment decisions.

Data from nine equity, index, and fixed-income funds were analyzed intraday with high frequency (daily and fifteen-minute intervals) from 2019 to 2022. Three main families of HAR models were evaluated by considering the relevant variables.

The results revealed that models based on second-order variations outperformed others in forecasting Realized Volatility (RV). Additionally, CVaR prediction was more accurate for index funds than for equity and fixed-income funds, with the HARQ model demonstrating superior performance.

Originality/Value: 

This study investigates the application of HAR models in predicting ETF risks and provides a novel framework for risk management and investment decision making, particularly in the Iranian financial market.

The present study aims to identify the most important variables affecting the fluctuations of gold prices. It is the first Iranian research in which the fluctuations in this market are modeled using non-linear Bayesian Model Averaging (BMA) and deep neural network approaches.

It is applied research where monthly data collected from 2010 to 2022 were used. It evaluates 35 factors playing a role in gold price fluctuations. GARCH and random fluctuation models are used to extract gold price fluctuations. TVPDMA, TVPDMS, and BMA models are used to identify the most important variables causing gold price fluctuations. Furthermore, the deep learning approach is used to investigate how effective the selected variables are in gold price fluctuations.

The results indicated that Support Vector (SV) models were more accurate than GARCH models in capturing fluctuations and that BMA outperformed TVPDMA and TVPDMS. Additionally, 12 variables were identified as influential in gold price fluctuations, with in-market factors playing a more significant role than out-of-market factors. The study also employed Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM), and Multi-Layer Perceptron (MLP) neural network models in deep learning mode to predict gold price fluctuations. It was concluded that global interest rates had the most significant impact on fluctuations in the gold price, with the Pivot Point DeMark's Index making the greatest contribution.

Originality/Value: 

The gold market is known for its volatility, and accurate predictions about its future can significantly impact decision-making. Understanding the gold price and making correct forecasts can help inform decisions about buying and selling gold in global markets, and determine the most favorable times for transactions and investments. Therefore, it is crucial to accurately predict the gold price from various perspectives. This research attempted to develop an intelligent model for forecasting fluctuations in the gold price.

In this research, a hybrid approach based on fuzzy logic and artificial neural network is presented to predict the failure of machines in order to increase productivity. The subject of this research is one of the factories of the automobile industry named Diaco Ide Aria, which operates in the field of automobile parts production. Preventive maintenance requires correct prediction of breakdowns and accidents, equipment and machines so that productivity can be increased by timely and correct maintenance of machines as well as fixing defects and breakdowns. To model the multi-layer perceptron fuzzy-neural network (MLP), first, 100 failures and stops were collected in a period of 15 months and then entered into MATLAB software. The obtained results show that the implementation of fuzzy-neural network and the prediction of machine failure time has reduced the duration and cost of repairs. Therefore, the working time and accessibility of the machines increased and ultimately increased the productivity by 57%, also, the accuracy of the developed neural-fuzzy model was estimated at 94%.

Today, maintenance and repair have become very important in the manufacturing industry. An ecient solution to prevent downtime is to predict equipment failure. Therefore, accurate and correct prediction of breakdown events in the eld of predictive maintenance can be very useful. In general, each prediction will be accompanied by a certain amount of error, which in various ways tries to control this error or limit it to a reasonable amount. In this thesis, a framework has been proposed that species when the system under review will need maintenance and repairs to prevent downtime as much as possible. Therefore, the main purpose of this thesis is to design and implement an ecient combination structure to accurately predict failure events using both standard statistical standard models and machine learning in predictive maintenance. The literature review results indicate that the use of these methods in recent years has led to extensive advances in the eld of providing accurate forecasts and subsequently improved the level of decisions made by managers and decisionmakers. The proposed model is used to predict failure events in benchmark data related to the truck air pressure system. Finally, the performance of the proposed model is compared with other data-driven techniques individually and in combination, which includes logit models, support vector machines, and multilayer perceptron neural networks. According to the numerical values obtained from the nal analysis, the results indicate that the backup vector machine model has higher prediction accuracy than other single models, and also the results indicate the eciency and eectiveness of the proposed parallel combination structure compared to the use of models individually and in series combination in modeling and forecasting issues. The parallel hybrid model improved the accuracy of predictions by an average of 11% in test data and 7% in training data. Therefore, due to the greater accuracy in combining classical statistical models and machine learning in parallel, the use of this combined method to improve the accuracy of predictions in the eld of predictive maintenance is recommended for future studies.

The most basic indicator of economic and social development in many developed and developed countries is the automotive industry of that country . ntil 2009, Iran was one of the top 18 automakers in the world and one of the largest automakers in the Middle East. In this research, the price of domestically produced cars in the market is predicted based on the dynamic system model. The statistical population is intended for collecting analytical data of 512 experts and specialists in this field, which according to Cochran's method, 412 people are considered as a statistical sample. The data obtained in the questionnaire were first extracted using the AHP method to extract the effect coefficients of the variables. Then, based on the dynamic system method, model design and simulation were performed in Wensim software. Given the current situation, if it is stable, the growth of car prices is low for 34 months, but after 34 months, a very significant growth can be due to the high impact of factors affecting car prices relative to reducing factors. Is the price of the car.

The bullwhip effect in the supply chain could lead to fluctuations such as extra inventory and delayed order. In the meantime, proper demand forecasting can significantly resolve these fluctuations by eliminating the bullwhip effect. The present study considers forecasting of retailer demand, suppliers and calculates the bullwhip effect in the two-level supply chain. Markov switching model and autoregressive model along with moving average are used to predict retail demand and supplier, respectively. The results showed that the lot size policy-based ordering system can reduce or completely remove the bullwhip effect. Besides, the bullwhip effect is appeared in both levels of the supply chain during utilizing the rm(q0,Qm) order-point policy-based ordering system.

Corona is currently the world's health crisis and the biggest challenge humans have experienced since World War II. Given the epidemic of the disease, it is invaluable to forecasting the number of cases and the resulting deaths to better understand the current situation and provide a short-term plan by managers. Accordingly, in this paper, a neuro-fuzzy network model is proposed to forecast the number of cases and deaths in countries that are most affected by this disease. The performance of the proposed neuro-fuzzy network has been compared with time series forecasting neural network as well as radial basic functions neural networks. The proposed model is able to predict the number of cases and deaths from the disease for a period of the next 15 days at a lower error rate.

Costumer information can be take an important basis in the analysis of costumer behavior. It's obvious that one of the most essential characteristics of the customer behavior is the customer risk modelling in order to the customer risk assessment. Today, it's very important to keep up stable costumers to confront with the rival market and get it. In this research, at the first we introduced introduction in order to survey prior research about risk modelling and assessment in work field. There are a lot of procedure to risk modelling because the usage of this methodology is very comprehensive whiles there is not any distinctive structure to risk assessment and modelling. In section two, we are considered descriptive variables of customer such as age, weight, usage, prehistory and occupation to analyze past behavior of customer with respect to the future behavior by definition experimental function from determinative historical data. It approach used to model customer future behavior. Then we are assessed purchase risk in order to predict the future behavior of customer. At first, according to the many characteristics that driven from the specific sample of new strongbox company customers, experimental functions generated and are compared to gather with the information that gains from the descriptive statistics and distribution diagrams on this data and then, purchase risk is evaluated experimentally. In the next section, the Bayes risk of customer is analyzed and used to classify customers according to the prior data. After it, we proposed guidance for improve the production programing and sale management decision tree technique. The approach mentioned in this research is used as a case study about the products of Kaveh strongbox company that readers can be realize the practical usage of this research as much. All data in this research that obtains from thoroughbred replier is done by expert questioner. The software that we used in this research are MINITAB and Expert Choice.

This study examines the efficiency of various topologies of deep learning networks (a superior approach to modeling and fitting socio-economic time series) in load demand forecasting using the data collected from a four-year period of households in Kurdistan City, Iran. Since the consumption pattern is a nonlinear and complex curve with a strong delayed dependency pattern, its prediction is not accurate by conventional statistical methods and the error reduction of this prediction has a significant effect on reducing production costs, unwanted squandering and fines. In this study, full-connected, recurrent, and also hybrid of them were investigated using the mean efficiency of absolute error percentage and mean square error index. When the input of the neural network was in the form of tensor, designing the structure of the deep neural network would be straightforward. In this case, the network can be implemented with a linear stack of layers sequentially. Although the sequential the sequential model is so common, it is inflexible when the input data is not in the form of tensor, e.g., Figure 4. Besides, in a forecasting model, each determinant might need a different type of neural networks such as CNN, LSTM or GRU. To overcome this challenge, we innovatively proposed parallel deep branches in our framework to represent the history of each determinant individually. The parallel branches process their determinants by using RNN and Dense networks. Then, the branches were merged together through concatenated and dense layers. The results indicated the superiority of the network topology as a combination of all connected and reciprocating models for modeling and predicting consumption. This superiority, due to the nonlinear nature of complexity, the strong attachment to the data of previous periods, and the existence of different degrees of delay in the exogenous variables of the problem can be fully justified. Considering that for excited peak load prediction, exogenous variables of the model (representing different atmospheric conditions) and artificial variables are included, this model has acceptable stability, compared to the models presented in previous studies.

The present paper presents the possibility of predicting firms' bankruptcy with Sprint, Altman, Fulmer, Zmijewski and Mckee Genetic models among companies in the Tehran Stock Exchange in a different way from previous research to introduce companies which have the potential for higher bankruptcy with a comparative approach among the models. To achieve this goal, 75 companies that are selected not covered base on 141 of the Commercial law. Required data for the 10 years (86-95) has been compiled. According to the results in each of the above models, some companies were identified as high-risk probability companies, and then companies that were identified as most likely to be bankrupt in most of these models. The results also show that, with the exception of Mckee model, in four other models, three companies with high bankruptcy probability were included. Among these four models, Zmijewski model has a higher coefficient of determination, hence we can say that relative to Other models have been more accurately predicted for bankruptcy and have a significant role in corporate bankruptcy among financial ratios, debt ratios, asset turnover, and asset returns.

One of the important issues that researchers and scientists face in the area of decision making and Predicting are choosing the variables that influence decision output and prediction. Therefore, if you can predict stock returns with appropriate variables and provide models for it, there will actually be more secure conditions in the capital market, which will help expand investment in financial markets. Therefore, the purpose of this paper is to examine the prediction of negative stock returns in Iran's capital market. In order to achieve this goal, data about 180 companies over the period 2010 to 2016 that was collected through a systematic elimination method that uses leverage, performance, turnover, volatility, quality, and Torpedo criteria to predict negative returns The results indicate that all of these criteria are all effective measures to predict the negative returns of shares in Tehran Stock Exchange.

The present research is aimed at predicting the beta coefficient (systematic risk) prediction dynamics within the framework of two macroeconomic structural models, the model in the context of dynamic stochastic general equilibrium (DSGE) and Panel Vector Autoregressive (PVAR) with the inclusion of financial data of companies and Some of the facts observed in the Iranian economy during the 15-year period (2002-2016). The results of the research show that economic shocks affect the beta coefficient of the stock. Also, in three approaches to predict stock beta coefficient, the VAR model has a lower error than the DSGE model. Finally, by comparing the moments of the present variables in the DSGE model and the real data of Iran's real moments, it shows the relative success of this model in the realities of Iran's economy.

Accurate monthly power demand network forecasting can help to plan the energy and it can handle the correct management of the power consumption. It has been found that the monthly electricity consumption demonstrates a complex nonlinear characteristic and has an obvious seasonal tendency. One of the models that is widely used to predict the nonlinear time series is the support vector regression model (SVR) in which the selection of key parameters and the effect of seasonal changes could be considered. The important issues in this research are to determine the parameters of the support vector regression model optimally, as well as the adjustment of the nonlinear and seasonal trends of the electricity data. The method that is proposed by this study is to hybrid the support vector regression model (SVR) with Fruit fly optimization Algorithm (FOA) and the seasonal index adjustment to forecast the monthly power demand. In addition, in order to evaluate the performance of the hybrid predictive model a small sample of the monthly power demand from Iran and a large sample of Iran monthly electricity production has been used to demonstrate the predictive model performance. This study also evaluates the superiority of the SFOASVR model to the other known predictive methods. In terms of the prediction accuracy, we used the evaluation criteria such as Root Mean Square Error (RMSE) and mean absolute percentage error (MAPE) as well as Wilcoxon's nonparametric statistical test. The results show that the SFOASVR model has less error than the other forecasting models and is superior to the most other models in terms of Wilcoxon test. Therefore, SFOASVR method is an appropriate option for prediction of the power demand.

After deregulation in electricity markets, huge amount of studies were done especially in designing new systems and energy pricing in order to improve efficiency of power systems and increase investors’ profit. Investment’s profit could be increased by better contracts and better price bidding for buying and selling energy in electricity market, as a consequence price forecasting is essential. The main goal of this paper is to predict price of electricity in Iran’s electricity market, using combined of fuzzy-neural network with Particle Swarm Optimization (PSO). In this paper, past prices, past loads, working and nonworking days, day hours and effect of seasons in 1394 have been taken into account as effective factors in forecasting mechanism. Combined model, is more precise in contrast to other methods like ARIMA,neural network, neural-fuzzy network and combination of fuzzy-neural with genetic algorithm. In following, process of price fluctuations has discussed for improving and increasing effectiveness of bidding. Results of simulation revealed that price forecasting is much more precise with price process mechanism.

Among various applications of time series, applying this concept in production industries with the intention of pre-detecting failure times of machines and implementing maintenance tasks, is considered as one of the most valuable activities in the automotive industries. With this regard, this article embarks on applying time series analysis as well as quality assurance concepts to detect failure times and implement proper actions. With this regard, we will analyze the data derived from maintenance department of the company. Then we determine influential factors on parts failure by means of quality assurance concepts.

In today's world of production, a large amount of information, including product and process design, assembling, material planning, quality control, planning, repair, maintenance, error detection, etc., is collected in database management systems and data warehouses. Therefore, the use of data mining in different areas of the production process has been growing dramatically in recent years. This paper reviews the studies on the discovery of knowledge and data mining applications as an important tool in the wider field of production. The purpose of this research is to provide a new framework for research efforts in relation to the current methods of using data mining in production based on the type of extracted knowledge, the used method and thus the identification of promising fields for study. The reviewed articles divide the application of data mining in the production process into four categories, including specifying, predicting, classifying and clustering. In this study, the applications of each tool mentioned in the different sections of the production of a product or service are introduced in order to expand the scope of future research.

Prediction of equipment remaining useful life (RUL) is essential for efficient maintenance decision making to decrease the maintenance cost. The failure history data and the expert knowledge are two important information sources for RUL prediction. Although there are lots of methods in literature that have used the history data to predict the equipment RUL, the hybrid methods has received less attention in this field. Therefore, this paper aims to present a new method based on a Takagi-Sugeno-Kang (TSK) inference system combined with information gathered from both condition monitoring process and expert knowledge to predict RUL of the equipment. In this paper the rule base for fuzzy inference system is prepared in two stages. At the first stage three basic rules are tuned with history data using a neuro-fuzzy (NF) network and in the second stage the rule base is completed by rules extracted under experts’ supervision. The performance of new hybrid method is evaluated in different real conditions to compare with traditional datadependent methods. Also, in this work a simulating algorithm is presented in order to generate different conditions that really could happen. Simulating parameters are estimated from real data related to bearing failures. The experimental results show that the efficiency of proposed method is higher than traditional data-dependent method.