جستجوی مقالات مرتبط با کلیدواژه « Prediction » در نشریات گروه « اقتصاد »

The unemployment rate and Gross Domestic Product (GDP) are among the most important economic indicators that understanding their current and future trends can help policymakers and decision-makers adopt appropriate solutions to prevent crises and improve the country’s economic situation. Accurate prediction of these two indicators can be useful in future planning and improving the country’s economy and people’s livelihoods. In recent years, artificial intelligence techniques and tools, given their many capabilities, can play a very important role in predicting important economic indicators. Therefore, given the high importance of the two indicators of unemployment rate and GDP on the economy of our country Iran, this article intends to first predict these two indicators separately and then predict the rate of GDP growth based on the unemployment rate using artificial intelligence techniques. For this purpose, in this research, seasonal data related to GDP and its components and the unemployment rate for the years 1976-2022 have been used. Also, machine learning models based on regression have been used for prediction. The results show that the predictions of the mentioned models have an appropriate accuracy in terms of evaluation criteria such as root mean square error, mean absolute error, mean absolute percentage error, which indicates.

In this research, using the method of fuzzy neural networks, the price of Bitcoin is predicted. In order to identify the appropriate criteria in this research in order to predict the price of Bitcoin, we have used previous studies and researches in this field in the first stage. In the following, using interviews with experts and experts in this field, the available information about Bitcoin became the final factors. Research information was collected using related sites and identified criteria. In this way, we first normalized the collected data. In the next step, by entering the normalized information into the MATLAB software and using the designed toolbox and using the fuzzy neural network method, Bitcoin price was predicted. In this way, 60% of the input data, which includes 1330 data, was considered as training data and 40% of the data, which is 887 data, was considered as testing. The research results show high accuracy prediction using the proposed method. As the error was considered in two cases, a small value was calculated for the error of the method.

Predicting stock prices is complicated; various components, such as the general state of the economy, political events, and investor expectations, affect the stock market. The stock market is in fact a chaotic nonlinear system that depends on various political, economic and psychological factors. To overcome the limitations of traditional analysis techniques in predicting nonlinear patterns, experts over the last two decades have used intelligent techniques, especially networking. They have used artificial neural networks and genetic algorithms to improve stock price forecasting. This study, considering the increasing development of forecasting methods in financial markets and also, since stock price is one of the most important factors influencing investment decisions and its forecasting can play an important role in this field, In this research, an attempt has been made to provide a model based on which the movement of the stock price can be predicted with high accuracy. Accordingly, a hybrid model for predicting stock price movement using artificial neural network is presented. For the statistical sample, the top companies of the stock exchange in the second quarter of 1399 have been selected. Then for each purpose, 32 variables were calculated. These variables are the input of the model and are optimized using the artificial neural network algorithm. The results show that the model performs much better in predicting the movement of stock prices and has a higher accuracy compared to traditional methods.

The purpose of this study is to Prediction the possibility of a banking crisis in a dynamic early warning system framework. We using data compiled from 10 middle-income countries for the period 1996-2017 and estimate static and dynamic logit model. The estimation results show that the dynamic logit model is better than the static model. We find that broad liquidity ratio, domestic credit to GDP ratio and stock market index are early warning signs of banking crisis. Also, the dynamic variable (lagged banking crisis) shows that if a banking crisis occurs in a year ago, there is a possibility of crisis in this year. Our result shows the possibility of continuation and persistence of banking crises for consecutive years. Then, the evaluation results of warning system show dynamic early warning system turns out to exhibit significantly better predictive abilities than the existing static one, both in- and out-of-sample.

 Predictive models for diagnosing bankruptcy or financial crisis have been widely discussed in studies and articles in the fields of economics and accounting and have been considered by financial institutions. One of the methods that can be used to help take advantage of investment opportunities and better allocation of resources is to predict financial distress or bankruptcy of companies. So, by providing the necessary warnings, can be alerted companies to the occurrence of financial distress so that according to these warnings they can take appropriate action, Secondly, investors and creditors can identify distinguish investment opportunities from unfavorable opportunities and invest in the right opportunities. Timely foresight can help decision-makers find solutions and prevent bankruptcy. The main aim of the current study is to express, determine and explain the predictive power of bankruptcy and profitability models of Tehran Stock Exchange companies to evaluate their performance and financial status by logistic regression using financial ratios selected by artificial neural network and Fulmer models.  

The method of the present study is applied in terms of purpose and descriptive in nature. Logistic regression technique was used to test the hypotheses. The results are presented in two parts: descriptive and inferential statistics. Collection of information from the financial statements of 132 companies of Tehran Stock Exchange during the years 2012 to 2018. Firstly, the initial classification and processing of information was performed and then Eviews software was used to fit the Fulmer model and Spss26 software was used for the neural network model. Suitable indicators based on the research background in the models include debt-to-equity ratio of shareholders, profit before interest and taxes, total liabilities to assets, receivable accounts ration to sale, net return on assets, long-term debt to assets, working capital, net profit to to sale.  

The research results indicates that both artificial neural network and Fulmer models have the ability to detect bankruptcy prediction with different accuracy, but the predictive accuracy of artificial neural network model is higher and has better performance compared to Fulmer model. In the artificial neural network model, the variables of working capital, receivable accounts on sales, net profit on assets, net profit on sales and long-term debt to assets are significant at high level in predicting corporate bankruptcy. Also, among the financial ratios used, the ratio of receivable accounts on sales had the most impact and the debt-to-equity ratio had the least impact on determining bankruptcy among the available variables.

The best way is to take preventive measures before the occurrence of financial incapability of companies and in this regard, the result of the present study confirms the use of artificial neural network method to predict the bankruptcy of listed companies. And also, the crtiteria of working capital, net profit on assets, ratio of total debt to total assets and net profit on sales are related to transactions with bankruptcy. That is, the higher the ratio of these ratios, the probability of bankruptcy is lower. Therefore, by issuing the necessary warnings to decision makers and as a result of their actions, companies can be guided in the right direction in order to avoid wasting resources.

Due to the complexity of the stock market in Tehran, the timing of transactions is very important. The timing of trading transactions helps analysts and traders to predict the stock prices movement. Therefore, the purpose of this study is to predict the timing of stock trading of listed companies in TSE.

The statistical population of the study consisted of all companies listed in TSE between 2013-2016. The sample is based on the systematic elimination method of 17 active companies in the TSE. Research method based on stepwise regression and fuzzy neural network based on indicators of relative strength, Moving Average Convergence-Divergence, simple moving average, Stochastic, EMA and Signal line.

The results showed that the average prediction accuracy of all networks created (96.55%) was more than random (50%). By applying the trading rules, the predicted values were converted to the signal It was suggested that the final signal of the designed system be obtained from the sum of the signals generated by the five technical indicators. Next, to evaluate the returns of the proposed transactions, the model AE using the trading strategy proposed study is a trading simulation assumptions were.

The efficiency of transactions made on the basis of the final signal proposed system efficiency methods, technical and purchasing methods and stored (in two cases before the deduction of transaction costs and after deduction of transaction costs ). Due to the positive results of SMA, EMA, SO and the proposed method, we can conclude that using these technical analysis indices in the Iranian stock market can predict the stock price trend. Meanwhile, the simple moving average method has the highest credit for predicting stock price trends. As a result, the Tehran Stock Exchange has the potential to apply various technical analysis indicators.

Today, special attention has been paid to the capabilities of chaos theories and neural networks and the application of these two models in financial markets, especially petroleum products. In this study, daily values ​​of stocks of Iranian petroleum products during December 2007 to June 2016 have been studied. Due to the nonlinear nature of financial data, chaos theory is used to study the chaotic amount of time series. The chaos theory based on Liapanov's exponent and fractal dimension studies the time series caused by nonlinear dynamic systems. In chaos theory, first, using the Lyapanov diagrams and the Poincaré surface of section map and measuring the correlation dimension, the possibility of chaos in time series of daily value of oil products in Iran is investigated. Then, the Lyapunov diagram is plotted using the delay time estimation obtained from the method of the average of mutual information and embedding dimension using the algorithm of the false nearest neighbors. The Lyapunov map and the Poincaré surface of section indicate a chaos in the investigated time series. According to the provocation of chaos in this time series, its nonlinearity was deduced. Therefore, a suitable neural network was designed and the best model was selected to predict the future prices of petroleum products stock with a correlation coefficient of 0.99831 and the error of the training data is 0.0012 and the error of the test data is 0.002 that indicating good accuracy in modeling the price of these industries and can be used to predict its future price.

The stock market agents should increase their prediction accuracy to maximize their returns, and it needs some advanced tools. In this article stock price of 50 companies in the Tehran Stock Exchange have been modeled using feedforward artificial neural networks. In this way, the daily stock prices are used from December 1384 to December 1394. The predictions accuracy are evaluated with four statistical indicators. The results show that accuracy of ANN predictions is very high. In some cases, although the prediction accuracy is higher, the correctness is lower. Therefore, in the assessment of the prediction, evaluation of the correctness has a significant contribution

The aim of this paper is to specify and estimate such a model for non-oil export that allows to revise the previous predicted volume of annual non-oil export, as soon as new seasonal data for the explanatory variables are released. This cannot be achieved unless a model is constructed in such a way that annual non-oil export is set as a function of some seasonal explanatory variables. By utilizing a very recent method of regression specification, namely MIDAS regression, this article sets annual non-oil export as a function of seasonal real GDP, seasonal real exchange rate and seasonal real exchange rate fluctuations. To provide more accurate predictions, as compared to traditional methods, this regression model is capable of providing a revised prediction as soon as new information concerning explanatory variables are released. The specified regression model, which is estimated by using time series data within the period 1989 – 2014, predicts the real amount of annual non-oil export for 2015 with a small error of almost 2%. So, this prediction is considered to be very close to the reality.

Energy as a driving force in most manufacturing and service activities plays an important role, and since the core of the economic growth process is GDP growth, it plays an important role in the economic growth and development of countries. In addition to energy price fluctuations as well as scarcity of energy resources, another important factor that doubles the need to examine the relationship between energy production and consumption is the environmental issues that all countries face.
An outlook analysis of energy consumption forecasts enables managers to take the necessary measures to control the variables of supply and demand for energy. This bottleneck helps provide policy advice to support investment. Increasing the accuracy of many energy consumption forecasting models requires identifying all the trends and variables and strategies affecting energy markets. Therefore, in the present study using gray systems theory (GM), gray prediction model is considered as a suitable model to improve energy consumption prediction performance. Also the energy consumption up to 1400 is predicted and estimated according to the model.

The savings becomes to invest in the capital market and then import into the production cycle and helps to the development and growth of countries. However, inefficient capital markets, cause savings to flow into real assets. Gold is a real asset, liquidity with high strength, and a suitable replacement for money. This wealth is a booming market in Iran. Fluctuations in the price of gold in addition to the influence of other markets can also affect other markets. Therefore, it is important for the state and the people to understand the trend in the price of gold and gold coins. The gold price forecast will help policymakers to make the right decisions. On the other hand, it is difficult and complicated to accurately predict the real variables. We need to recognize the structural nature is predictable pattern. In this article, chaos theory was used to identify the structural nature of the time series of Bahar Azadi gold coin.
Theoretical Framework: Chaos theory analysis of the systems that have non-linear relationships and irregular time series. Economic time series variables follow a stochastic process and thus are not predictable. However, the series are not random, and are expected in the short term. There are tests for chaos in time series, such as correlation dimension, BDS, and Lyapunov exponent maximum test. Results of the study by Kim et al. (2003) showed that the BDS test is more efficient than other tests.
For the purpose of this study, the non-linearity of the BDS test, and the Lyapunov exponent maximum test of the chaotic time series were used. BDS test was conducted in three stages: the original data, the residual of ARIMA, and the residual of GARCH. To determine the structure of time series of Bahar Azadi gold coin, 1670 observation was divided into 8 groups of the two hundred. Null hypothesis test is the IID and independent data. The Lyapunov exponent maximum test check on all data. Positive values of the statistics indicated the existence of chaos in the system. R and MATLAB software were used for data analysis.
First, the stationary data were checked. Dickey-Fuller test the null hypothesis is accepted, which implies the existence of a unit root. The first stage of BDS test was performed on the original data in the dimensions inscribed. The results showed that the null hypothesis was rejected, except the first group. As a result, the original data were not IID, and linear or non-linear dependence exists between them. Before the second phase of the test, the appropriate ARIMA model was selected. The unit root test was performed on the residual of ARIMA, and the null hypothesis was rejected. As a result, BDS test was conducted on the residual ARIMA. In the third stage, first the variance heterogeneity was checked, white test the null hypothesis is rejected, thus confirming the heterogeneity of variance. Then, the existence of ARCH effect was checked. ARCH effect in the first five groups, GARCH effect in the next three tests by Ljung-Box and LM-ARCH was confirmed. According to the BDS test conducted on the residual of GARCH, the null hypothesis was rejected, which residual IID, and linear and nonlinear dependence does not exist, thus confirming the process of chaotic time series data structure of Bahar Azadi gold coin. Wolfe algorithm was used in this study to calculate the Lyapunov exponent maximum test. The results showed that the Lyapunov exponent was small and positive for all aspects and intervals.
As a result, time series of Bahar Azadi gold coin is possessed of a chaotic process. So we can predict future prices with the non-linear model in this series.

In many studies, in order to predict economic variables, quantitative methods based on time series data or cross-section data are mostly used. Time series data or cross-section data do not control the heterogeneity of countries, and the possibility of obtaining the risk of biased results exists. Panel data provides more informative data and a more degree of freedom which lead to results and predictions that are more precise. In this study, while considering the significant role and proportion of dried fruits in non-oil exports, the synthetic artificial neural network-panel data method has been used to predict the price of pistachio, raisin and date exports. Then the predictions were compared, using the accuracy criteria, with the regression model (the two-way error component model). The data from ten target markets for each of the dried fruits from 1992 to 2012 were used, and the results of this study show that the new and synthetic artificial neural network-panel data method has a better performance in predicting the price of Iran’s pistachio, raisin and date exports than that of the regression method. Therefore, it is suggested that exporters, policy makers, and researchers use this method in order to predict economic variables.

The relationship between energy and agricultural products prices is an important and influential factor on food price surge. On the other hand، data availability in different frequencies is a dilemma facing time series econometricians، because by averaging of the data some valuable information in high frequency data will be lost. MIDAS regression models have recently been developed as an alternative dealing with mixed frequency data problem. This study applies generalized ARDL approach to estimate MIDAS regression for prediction of cereal prices using quarterly data on exchange rate and annual data on energy prices، interest rate، and inflation for the period 1982-2008. Prediction accuracy Statistics show that MIDAS model provides more accurate prediction of cereal price compared to simple averaging method.

As one of the important energy forms، natural gas consumption has an upward trend in recent years. Therefore management and planning for provision of it requires prediction of the future consumption. But many of prediction procedures are inherently stochastic therefore it is important to have better knowledge about the robustness of prediction procedures. This paper compares robustness of two prediction procedures; Artificial Neural Networks as a nonlinear and ARIMA as a linear model. using resampling method to predict the monthly consumption of natural gas in the household sector. Data spans from 2001-4 to 2012-3، to train the networks، we used genetic algorithms and Particle Swarming Optimization then results were compared using 10-fold method. According to the results، the particle swarm optimization (PSO) outperforms the genetic algorithm. Then we used data from 2001-4 to 2010-3، with resampling by 2000 to predict the natural gas consumption for the 2001 -4 to 2012-3 and to form critical values. Results show that prediction by a mixed method using ANN and PSO is more robust than ARIMA method.

during previous years, Iran’s economy has been faced with numerous fluctuations, while in some cases it turned to a financial crisis as that of 2011-2012. However, the main question is that whether the financial crisis during 2011-2012 has been similar and predictable based on previous financial crises of Iran’s economy. In the present research, firstly, based on previous studies in the subject of Iran’s financial crises, the years of financial crises in Iran’s economy during years of 1978-2012 were determined. The corresponding results have indicated that three kinds of financial crises including Currency Crises, Money Crises and Twin Crisis (simultaneous Money and Currency Crisis), were occurred in Iran’s economy during 1979-2008. Then the financial crisis periods (in which two or more financial crisis years are included) were determined, the political and economic events during each of them were expressed and the reasons of occurrence of financial crises in each of them were described. Then the data of 10 macroeconomic variables of Iran’s economy belonging to 1978-2012 were deployed. Based on data of 1979-2008 as input data and information about crisis and non-crisis year and special kind of financial crisis in each crisis year as target data, the LVQ model was trained, in order to learn crisis and non-crisis patterns (behavioral patterns of macroeconomic variables in crisis and non-crisis years), and to learn to recognize the various kinds of crisis patterns (to understand which kinds of financial crises has been occurred in each crisis year). Then the trained model (trained model based on data of 1979-2008) was simulated using the data of 10 macroeconomic variables of Iran’s economy during 2008-2012, in order to access the capacity of occurred financial crises of years of 1979-2008 to explain and predict the financial crisis of Iran’s economy during 2011-2012. The results of simulation process indicated that the financial crisis of Iran’s economy during 2011-2012 has had a completely different kind of crisis pattern, compared with those patterns which have occurred in Iran’s economy during 1978-2012.

During previous years, Iran’s economy has been faced with numerous fluctuations, while in some cases it turned to a financial crisis as that of 2011-2012. However, the main question is that whether the financial crisis during 2011-2012 has been similar and predictable based on previous financial crises of Iran’s economy. In the present research, firstly, based on previous studies in the subject of Iran’s financial crises, the years of financial crises in Iran’s economy during years of 1978-2012 were determined. The corresponding results have indicated that three kinds of financial crises including Currency Crises, Money Crises and Twin Crisis (simultaneous Money and Currency Crisis), were occurred in Iran’s economy during 1979-2008. Then the financial crisis periods (in which two or more financial crisis years are included) were determined, the political and economic events during each of them were expressed and the reasons of occurrence of financial crises in each of them were described. Then the data of 10 macroeconomic variables of Iran’s economy belonging to 1978-2012 were deployed. Based on data of 1979-2008 as input data and information about crisis and non-crisis year and special kind of financial crisis in each crisis year as target data, the LVQ model was trained, in order to learn crisis and non-crisis patterns (behavioral patterns of macroeconomic variables in crisis and non-crisis years), and to learn to recognize the various kinds of crisis patterns (to understand which kinds of financial crises has been occurred in each crisis year). Then the trained model (trained model based on data of 1979-2008) was simulated using the data of 10 macroeconomic variables of Iran’s economy during 2008-2012, in order to access the capacity of occurred financial crises of years of 1979-2008 to explain and predict the financial crisis of Iran’s economy during 2011-2012. The results of simulation process indicated that the financial crisis of Iran’s economy during 2011-2012 has had a completely different kind of crisis pattern, compared with those patterns which have occurred in Iran’s economy during 1978-2012.

Stockholders for making proper investment decisions need information which assists them in making the best decision. Among available information، Information related to the per share forecasted earnings is the important ones in the user’s opinions. Beside، companies try to forecast the earnings per share with maximum accuracy to attract investors. Accordingly، the present study seeks to provide a model to improve earnings per share forecast of companies listed in Tehran Stock Exchange (TSE) using modern artificial neural networks approaches. For this purpose، first the factors affecting future earnings per share were inferred from internal and external research، then using the sample companies'' financial information in the years 2005-2012 and employing support vector machines and artificial neural networks methods، earnings per share forecasting was designed. Support vector machines model was able to forecast the sample companies’ next year’s earnings per share with an adequate error of 5%. This model introduces current year’s earnings per share with effective coefficient of 25% as most effective variable to forecast next year’s earnings per share. The results suggested that the support vector machines model has similar performance in comparison with artificial neural networks model.

In this research the comparative prediction of Iran''s banking system (included 14 banks) was carried out by using econometric and artificial neural network models. Accordingly، at first، by using the Kohonen neural network model، the considered banks were divided into two categories of high performance and low performance groups and then using the output of Kohonen neural network model، financial proportions and Panel Data econometric model، the performance of Iran''s banking system was estimated for the period 2004-2010 and finally by using models evaluation criteria، the performance of Panel Data and ANN models was compared. The results of Kohonen neural network model indicated that from 14 considered bank، 4 banks belong to high performance group and 10 banks are belong to low performance group. Also the results of Panal Data estimations showed that “capital income/total income «portion has the lowest and “cash/total deposits»، has the haighes effect on the Iran''s banking system. Finally the results of models comparison stated that the ANN model outperforms the Panel Data model to predict the performance of Iran''s banking system.

This article aims at presenting a descriptive study of the future of Iran’s economy. The most significant feature of the article is that it taps into the new institutional theoretical framework and compares it to a scenario-based model. On this basis، technology is the key concept that explains the reality of development، and the efficiency of ownership law structure is the cause of success of a national economy in the realm of technology. This article attempts to present a picture of Iran’s future economy within the new institutional theoretical framework، through the necessities associated with changes in the structure of ownership law along the path of institution changes. The scenario-based model assumes that any changes in the structure of the ownership law aiming at enhancing technology require that the national economy be able to pay the costs of institution changes، costs that result from an ascending efficacy of institution matrix. In other words، the winners of the institution matrix must be able to pay such costs. In this way، the efficiency of judicial system and political and economic control of the military is the yardstick of the movement. It appears that trends in Iran’s economy are in such a way that the costs of changes in the structure of ownership law aiming at enhancing technology are increasing.