جستجوی مقالات مرتبط با کلیدواژه "پیش بینی" در نشریات گروه "مالی"

Reliability is considered as one of the most important qualitative features of accounting information. According to the theoretical concepts of financial reporting, reliable information is free from errors and biases. Because of reporting scandals, the issue of financial statement restatement has garnered significant attention in accounting literature and journals. The main goal of the research is to provide a suitable criterion for predicting the re-presentation of financial statements using Benish models and Benish development.

This research employs an applied method and, in terms of methodology, follows an ex post facto approach. Also, this research falls within the realm of accounting evidence and is based on actual data from the financial statements of companies. The required data was extracted from the Codal Exchange website and from the published financial statements and reports of companies listed on the Tehran Stock Exchange during the period from 2008 to 2019. Furthermore, SPSS and Python software were utilized in this research to analyze the data.

In this research, firstly, Benish's primary model was examined to predict the possibility of financial statement renewals, and then, using indicators that affect renewals, independent t-tests, and step-by-step regression were used to estimate the model using logit regression. became. The overall accuracy of Benish's initial model for predicting the probability of resubmission of financial statements was estimated at 29.43%. The overall accuracy of the model regarding the developed model of Benish was estimated at 68.68%. The research results indicate that, based on the confusion matrix, among the predictive models for financial statement restatement, the Developed Benish model with logistic regression achieves a total forecasting accuracy of 68.68%, demonstrating the highest predictive power compared to the original Benish model, which has an accuracy of only 29.43%. Therefore, the research hypothesis is confirmed and the accuracy of the developed Benish model with logit regression is more than the original Benish model in identifying renewed companies.

The correct flow of information in the capital market, as the driving engine of the economy, leads to correct and rational decision-making by the participants and finally brings economic development and improvement of social welfare. Financial reporting serves as a primary means of communication between a company and various stakeholder groups. Therefore, information transparency and the provision of correct and timely information by companies admitted to the stock exchange is one of the most important effective factors in achieving an efficient capital market. This research seeks to provide an optimal model using Benish's profit manipulation model to predict the probability of the re-presentation of financial statements. Several studies have been conducted on this topic; however, there is a notable lack of empirical evidence regarding the development of an efficient and useful model tailored to the economic conditions of Iran's capital market for predicting the likelihood of financial statement restatements. In this study, firstly, the primary Benish model was examined to predict the probability of re-presenting the financial statements, and in the second step, to present a developed model based on the Benish indices and using logit regression, eight financial indicators of the Benish model from the financial statements of the company were examined. The selected ratios were extracted, and the stepwise regression test was applied to identify the best ratios for predicting financial statement restatement. In this context, the two final financial indices chosen for the selection and development model were the total accruals index to total assets ratio and the debt ratio index. Benish's findings were presented based on two final indicators. In the third step, based on the clutter matrix and comparing them, the best model and method for prediction was selected.

Due to the high costs of financial distress and bankruptcy and the importance of timely prediction of financial distress to take corrective actions, this study examines the research related to the prediction models of financial distress and bankruptcy in Iran and other countries. This study is a review of research that examines 102 research in Iran and 298 research in other countries during the years 1930-2023, describing theories and analyzing the nature, trend, composition, and periods of using models and presenting practical and research recommendations. The results indicate an upward trend of research in Iran, while it is declining in other countries. The noteworthy point is the use of various combinations of models and techniques in other countries is more diverse than in Iran, and the prominence of intelligent models that are based on evolutionary calculations. The research concludes by providing executive and research recommendations. The recommendations include diversifying predicting techniques, emphasizing models based on evolutionary computing algorithms, paying attention to the predictions after the crisis, developing special models for small and medium-sized companies, separating the criteria for recognizing financial distress and bankruptcy in the Tehran Stock Exchange, making dynamic predictions, and developing more focused and specialized models to increase accuracy. This study addresses the shortcomings of past research, provides more up-to-date information and insights on the dynamic predictions, compares these studies in Iran to other countries, and provides some executive recommendations and research topics based on the results.

Financial distress indicates that a company is approaching bankruptcy and the vast and profound negative effects of bankruptcy on society, necessitate reviewing bankruptcy prediction models. Therefore, predicting financial distress in its early stages can inform the stakeholders of companies about their future possible losses (Zhou et al, 2023). Global crises like the 2008 financial crisis and the COVID-19 pandemic have forced even very strong international companies to continuously monitor their financial situation (Woodlock & Dangol, 2014; Hassan, 2022; Papik & Papikova, 2023). These environmental dynamics require reforms and more accurate predicting methods for the financial health of enterprises, (Brygala, 2022) especially for small and medium-sized companies, which have weaker financial resources (Ciampi et al. 2021; Mirza et al., 2023). While existing literature reviews are often outdated or limited in scope, this study addresses these gaps by comprehensively analyzing the models and techniques used for predicting financial distress and bankruptcy in Iran and other countries from 1930 to 2023. Furthermore, the study compares the nature, trends, composition, and periods of use of predicting models in Iran and other countries and offers insights for future research.

Systematic review refers to examining, criticizing, and evaluating a specific research topic, extracting and interpreting data from published articles, and analyzing, and describing results based on clear evidence (Lasserson et al., 2020). It generates highly credible, low-bias, and quality scientific research documents (Yetley et al., 2016). This meta-analysis research systematically evaluates financial distress and bankruptcy prediction models by analyzing 102 studies from Iran and 298 from other countries from 1932 to 2023. Data were sourced from reputable Iranian and international databases and journals, including Emerald, Science Direct, and ProQuest. The study aims to rectify previous shortcomings, update information, and offer new research avenues based on the findings of this review.

In Iran, financial distress prediction studies that started in 2001 shows a rising trend, encompassing 13 studies conducted during 1968-2007 and 89 studies during 2007-2023. Financial distress prediction research in other countries started in 1932 and has shown a declining trend in recent years including 10 studies conducted during 1932-1968, 217 studies in 1968-2007, and 71 studies during 2007-2023. The 102 predicting techniques in Iran during 2001-2023 consisted of 43 statistical models, 40 intelligent models, 17 combined models, and 2 other models such as Data Envelopment Analysis and judgmental, etc. However, 298 studies in other countries include 117 statistical models, 125 intelligent models, 25 combined models, 21 theoretical models, and 10 other models such as Data Envelopment Analysis and judgmental, etc. Among the 102 papers from Iran, 61.76% introduced modified or new models, while 87.5% of global research presented innovative models. Techniques used globally from 1930 to 2023 include statistical, intelligent, hybrid, judgmental, and other models. Statistical models dominated earlier, however, intelligent models, particularly those inspired by animal collective intelligence, gained prominence and attention for their high accuracy.

Iranian research on predicting financial distress and bankruptcy exhibits an upward trend, while global research has been declining recently. The results verify that intelligent models like neural networks and genetic algorithms, notably those inspired by animal collective intelligence such as the ant algorithm, and the firefly algorithm, the bird algorithm, demonstrated higher prediction accuracy than statistical models such as multiple discriminant analysis, logistic regression, probit, and theoretical models. Successful intelligent models, which widely used recently and exhibited a higher accuracy, include gradient boosting models (Jones, 2017), machine learning methods (Chen et al., 2023), and models based on collective intelligence such as the firefly algorithm (Bayat et al., 1997). Among the statistical techniques, the nonlinear logistic regression techniques demonstrated a high level of accuracy (Lohmann et al., 2022; Lohmann & Mollenhoff, 2023). In addition, 61.76% of 102 models from Iran and 87.5% of 298 articles abroad were innovative models. Recommendations include diversifying predicting techniques, emphasizing models that are based on evolutionary computing algorithms, attention to the necessity of post-crisis predictions, developing tailoring models for SMEs, separating criteria for financial distress and bankruptcy recognition in the Tehran Stock Exchange, the necessity of adopting dynamic predictions and (Lohmann & Molenoff, 2023) developing focused and industry-focused models for enhanced accuracy (Nazmi Ardakani et al., 2017).

In the era when companies are facing many challenges to survive in competitive markets, it is important to identify the factors affecting financial crises. One of the ways to help investors and companies is to provide models to predict financial distress. The purpose of this research is to investigate the ability of rough set theory and compare it with artificial and fuzzy neural networks to predict the financial distress of companies active in the Tehran Stock Exchange. For this purpose, 329 distressed companies from non-distressed corporations were selected during the period of 2006 to 2020. The neural networks investigated in this research are: Multi-layer Perceptron (MLP) neural network, Radial Basis Function (RBF) network, and Adaptive Fuzzy Inference Network (ANFIS) as well as the software used to create the ROSETTA rough set and the software The tool used to design artificial and fuzzy neural networks is MATLAB software. The results obtained in this research show the high efficiency of the rough sets model with 98/7 percent accuracy for predicting financial distress.

Fama (1970) showed that stock markets have weak efficiency and follow the random walk model, so investors cannot achieve abnormal returns by using historical data. It is very important, therefore, to know about the stock price process. The refining and petrochemical industry companies’ stocks are among the most popular ones in the Iranian stock market because they often distribute appropriate dividends to shareholders and sometimes even have good price returns. Also, petrochemical companies are known as the leaders of the stock market, with a great effect on the main stock market index in Iran (TEPIX). This article is to test the random walk hypothesis or weak efficiency of daily stock prices in six petrochemical and three refining Iranian companies.
 
To test the random walk hypothesis, in the first stage, augmented Dickey-Fuller (ADF) unit root tests were conducted using the approaches proposed by Dolado et al. (1990) and Hamilton (1994), along with the Zivot and Andrews (ZA) unit root test incorporating an endogenous structural break. According to the first stage, GARCH (1.1) and Exponential GARCH approaches were used in the second stage to control the fluctuations and leverage effects in evaluating weak efficiency. Daily stock price data (adjusted) were used to test the weak efficiency hypothesis.
 
The results of the augmented Dickey-Fuller (ADF) unit root test and the Zivot and Andrews endogenous structural break showed that Iranian companies of Nouri, Parsan, Pars, Tapico, Shepna and Shetran are pure random walk (weak efficiency). However, Fars, Shepdis, and Shabandar follow a random walk with drift, suggesting the absence of weak efficiency in these companies. In addition, the results of GARCH and exponential GARCH models showed that there is a positive relationship between risk and return for all seven companies. Also, volatility shocks in Fars, Nouri, Pars, and Tapico companies are completely permanent (weak performance). In addition, the shocks observed in Parsan, Shepdis, and Shatran companies are transient, with their effects dissipating over time, and the prices readjusted to the long-term mean, indicating the absence of weak efficiency. The evidence confirms that in these companies, the volatility caused by negative (adverse) news is more than the volatility caused by the same level of positive (favorable) news.
 
According to the findings of the two stages, Nouri, Pars, Tapico, and Shapna companies have weak efficiency which means that the stock price behavior of these companies cannot be predicted. On the other hand, Parsan, Shapedis, Shatran, Fars and Shabandar companies show weak efficiency which means that their stock price behavior is predictable. The results of this article have important implications for investors in the Iranian stock market. Market participants can engage in effective modeling for stocks that lack efficiency and exhibit predictable behavioral patterns to some extent. This allows them to gain insights into future prices and potentially earn profits.

In recent years, in the financial literature, more attention has been paid to the level of cash holding of companies. So; Forecasting is important to determine the optimal level of cash holding. In this research, using linear and non-linear methods and 13 influential input variables, the amount of cash in 103 companies admitted to the Iran Stock Exchange during the years 2013 to 2021 has been predicted. The methods used include multiple linear regression (MLR), k nearest neighbor (KNN), support vector machine (SVM) and multi-layer neural networks (MLNN) for prediction. The results show that the traditional method of multiple linear regression has not been successful in predicting cash, but machine learning algorithms have been superior with an accuracy of 0.99. The variables of profit per share, the ratio of current assets to current liabilities and the ratio of short-term debt to total assets have had a greater impact in all algorithms. Therefore, managers can use advanced machine learning algorithms to predict the future cash flow of companies.

Profit management has been one of the most controversial topics in recent research. Most research on earnings management has examined the linear relationship between independent variables and earnings management using statistical methods but they did not use these variables to predict earnings management. Today, with the growth of information technology and the introduction of artificial intelligence, including artificial neural networks into the field of scientific research, it has become possible to study nonlinear relationships between variables. In this study, an attempt was made to estimate optional accruals for predicting earnings management using artificial neural networks. Also in this research, the genetic algorithm optimizer model and Particle swarm has been used to optimize the weights of the artificial neural network model to enhance the predictive power. Then, the ability to predict profit management was evaluated using the modified Jones statistical model, artificial neural network and the combined technique of genetic algorithm, Particle swarm and neural network. The sample used in this study included 150 companies listed on the Tehran Stock Exchange between 2015 and 2020. Findings showed that the artificial neural network has a high ability to predict profit management, compared to the modified Jones linear model. The findings also indicate that the accuracy and ability of the combined model of genetic algorithm, particle swarm and neural network in predicting profit management is higher than the combined model of genetic algorithm-artificial neural network.

Predicting time series of financial markets is a challenging issue in the field of specialized studies of time series and has attracted the attention of many researchers. Due to the presence of big data, this issue has led to the growth of developments in the field of machine learning models. Due to the importance of this issue, in this study, by using the comparison of different machine learning models such as random forest approaches, support vector machine, artificial neural network and deep learning-based recurrent neural networks to investigate the ability of different machine learning models in prediction. The total index of Tehran Stock Exchange during the period 2013 to 2020 has been discussed. The prediction results of 1, 3 and 6 day courses for the out-of-sample period show that the machine learning method based on the long short-term memory (LSTM) network, a recurrent neural networks, has a better result compared to other models.

It is very important to distinguish how the volatility in the return of assets occur. For this reason, in recent years, realized volatility and frequencies of daily volatility recognition studies have been developed. This study uses stock prices of 30 big companies of Tehran Stock Exchange during the years 1390 (2011) to 1394 (2016) and calculates the realized stock volatility during trading days using the HAR-CJ model to examine the effect of size and intensity of price jumps in predicting index volatility. The results showed that the development of HAR-CJ and HAR-RV-CJ models using the size and intensity of jump did not have a significant effect on improving the index volatility prediction but, to a small extent, the model prediction performance Adjusts for index volatility. Also, using intraday jumps instead of daily jumps, does not improve the performance of the prediction model.

One of the most important causes of banking system problems is not paying enough attention to credit risk. Ignoring the creditworthiness of applicants and mandatory facilities has left banks with a significant amount of problem assets and a large number of non-current facilities and has severely reduced their ability to finance. Intelligent methods such as multivariate switching model, due to the behavior based on multiple analysis, make it possible to find appropriate answers to predict the amount of risk. The purpose of this study is to model the credit risk of some Iranian banks using the Markov-Switching method. In order to prevent potential losses of banks, it is important to study the variables that have a significant impact on creating high-risk and critical conditions. This can be modeled through the Markov dual-regime model, which is a useful way to describe the process by which variables escape in a series of states at a continuous time being measured. Therefore, in this paper, by analyzing independent variables such as micro and macroeconomic variables, financial factors, external shocks and financial distress index using multivariate switching method to identify, score and determine the impact of each of the independent variables in credit risk control and consequently credit risk forecast is considered in banks. In this regard, three hypotheses were determined and the annual data of the member banks of the Tehran Stock Exchange in the period 2011 to 2020 were used to test the hypothesis. The results of experiments in MATLAB software show the proper performance of risk prediction accuracy of the method based on multivariate Markov switching.

Liquidity in the stock market expresses how close the stock is to cash. Since the Tehran Stock Exchange is included among the world's non-cash stock exchanges and the issue of stock liquidity is one of the main concerns of investors, therefore, in this research, an attempt is made to predict the liquidity of the stock exchange using deep learning models. The statistical population includes companies active in the Tehran Stock Exchange in the years 1394-1400, and 23 companies were studied as a sample. The transaction volume and value, stock turnover ratio, Amihood, the difference between the bid and ask prices (spread) and the relative spread were measured as liquidity measure and a fully connected neural network based on multilayer perceptron (MLP), mixed deep learning (MDL) model and classical linear regression (LR) model was tested. To measure the predictive power of the models, the mean squared error (MSE) and mean absolute error (MAE) measures were calculated, and t-test was used to compare the accuracy of forecasting methods. According to the results, the prediction error rate of MDL model is lower than the other two models, and the statistical tests also confirm the significance difference in the prediction accuracy of the models at the 95% confidence level, which shows the proper performance of the proposed hybrid model compared to two other models.

The capital market is growing rapidly. This has led to higher demand for information, more effort to predict and invent new models for predicting the future of the market. Predictive models can be classified into three categories. The first group uses technical analysis, the second group uses fundamental analysis, and the third group uses data mining and machine learning. In the present study focusing on data mining method to compare the accuracy of selected machine learning models including neural network, logistic regression, k nearest neighborhod, support vector machine and cross validation to predict stock prices for 12 selected companies of Tehran Stock Exchange that They have been selected through systematic deletion method in the form of machine learning models and the results of this paper showed that among the machine learning algorithms, the support vector machine algorithm has the highest predictive power in stock prices.Keywords: stock exchange, forecast; Stock prices, algorithms, machine learningJEL Classification Code: C8, G1

One of the most important qualitative features of accounting information is its reliability. According to the theoretical concepts of financial reporting, reliable information is free from mistakes and biased tendencies. As a result of the reporting scandals, the subject of restatement of financial statements has received much attention in accounting texts and journals. The aim of this research is to provide a suitable criterion for predicting the re-presentation of financial statements using selected financial and non-financial ratios of statistical tests in companies admitted to the Tehran Stock Exchange between 2009 and 2020. For this purpose, 265 companies were selected according to the restrictions. Also, in accordance with the research literature, financial and non-financial ratios have been extracted and classical logit model and meta-heuristic models of neural network and genetic programming have been used to estimate the forecast. The results of the research indicate that, based on the confusion matrix, among the models predicting the re-presentation of financial statements, the efficiency of the presented model due to the combination of financial and non-financial ratios (combined model) with the genetic algorithm is 85.66 percent accuracy. It has a total nose that has the highest predictive power compared to the model presented with logit regression.

Some studies have shown that investors' emotions can influence their financial and investment decisions, as well as stock returns; On the other hand, stock risk can also affect this relationship; Therefore, the present study has examined the moderating role of corporate governance system on the relationship between investors' emotional tendencies and changes in stock returns of companies listed on the Tehran Stock Exchange. In this study, the independent variable is investors 'feelings and the dependent variable is stock return fluctuations. To test the research hypotheses, the research model estimation method in the form of panel data and analysis based on the model of fixed effects and random effects by Eviews software based on composite data has been used. The sample includes 139 companies listed on the Tehran Stock Exchange in the period 1394-1399, which have been selected by a systematic elimination method. Findings showed that there is a significant relationship between investors' emotional tendencies and fluctuations in stock returns. Also showed that the variable of institutional shareholders can moderate the relationship between investors' emotional tendencies and fluctuations in returns.

The purpose of this study was to compare the efficiency of machine learning models (32 models) and statistical models (14 models) in predicting the financial risk of listed 145 companies in Tehran Stock Exchange during the period 2010 to 2020 and selecting the best model using advanced optimization techniques. Findings of the research using the test of comparing the accuracy of prediction coefficients, indicates that with 99 percent confidence, the prediction accuracy of machine learning models is higher than statistical models. Also, the best machine learning model after optimization was the evolutionary support vector machine model with 99.86 percent prediction accuracy and the value of the area under the curve was 0.998. In addition, accrual financial ratios with 99.45 percent predictive accuracy and operating financial ratios with 98.62 percent predictive accuracy were able to perform better than other financial ratios in using the evolutionary support vector to predict financial risk. on the other side, the projected financial risk varied according to different industries. Therefore, it was found that machine learning models can be used as an important tool in predicting corporate financial risk due to the lack of limitations that statistical models face.

The behavioral characteristics of managers are among the most crucial factors in making decisions on financing and capital structure of the company. The most important characteristics of managers include optimism, overconfidence, and myopia. Thus, the study investigated the influence of the behavioral characteristics (optimism, overconfidence, and myopia) of managing directors on the financial flexibility of companies with the moderating role of managing directors efficiency and political connections. Accordingly, the data of 130 companies listed in the Tehran Stock Exchange from 2014 to 2018 were examined using multivariate regression. Hypothesis testing revealed that there was no significant relationship between managing directors optimism and the financial flexibility. However, the relationship of managing directors overconfidence and managing directors myopia with financial flexibility was significant and negative. managing directors efficiency moderated the relationship between managing directors overconfidence and financial flexibility, whereas the political connections moderated the relationship between managing directors myopia and financial flexibility.

Predicting crises and price jumps in the stock market and based on different models has been growing over the last decade. Due to the presence of big data, this issue has led to the growth of developments in the field of machine learning and deep learning models. Due to the importance of this issue, This study examined the ability of different machine learning models to predict the jumps in the total index of the Tehran Stock Exchange during the period 2013 to 2020. For this purpose, first stock market jumps were extracted based on the ARJI-GARCH approach and then these jumps were predicted by considering the possible effective variables including global and domestic markets. The prediction results of 1-, 3-, and 6-day periods for the out-of-sample period show that the machine learning method based on the long short-term memory (LSTM) network, a recurrent neural network, has a better result than other models.

In today's growing, saturated and competitive markets, the insurance industry, like other industries, is suffering from customer churn. Recent advances in communications and technology have led to increased awareness and ease of comparison of insurance policies and services. Thus, insurers are constantly confronted with new offers from competing companies and easily turn to competitors; in other words, they churn. The aims of this research are actually pursued through the following research question. What are the main factors of churn in the Iranian insurance industry? This study is conducted in two phases: qualitative and quantitative. First, in the qualitative phase, 15 experts in the insurance industry are interviewed to identify the factors affecting churn. Then, based on the identified factors in the research literature and comparing them with the results of the interview, 8 main influential factors are identified and finalized. In the quantitative phase, these indices are first weighted by hierarchical analysis technique. Then, the statistical population is determined in the quantitative phase; they included the insurers of the Iranian insurance company, and by selecting a sample of 120, neural network technique is applied to fit the model. The calculated R = 0.74 proves that the eight identified factors (type of insurance, premium, final result of claims, duration of cooperation, payment method, number of installments, number of policies, and number of claims) can best explain the reasons of churn of policyholders.

Finding the best way to optimize the portfolio is one of the concerns of activists in the investment management industry. In recent years, the introduction of economic and mathematical models in the prediction of Gold indice has helped many investors to optimize portfolios. Therefore, in this study, we introduce models of GARCH family composition and convoultional neural network to predict the daily yield of Gold index will be paid during the period of 1390-1398. In this study, the Gold index is examined using GARCH and EGARCH short-term memory models. Of the two variables, the price of crude oil and the dollar index as factors that their shocks and fluctuations have a major impact on Gold indices are used as control variables. In addition to using convolutional model, considering the better performance of combined models (compared to individual models ) In anticipation In this study, all models of the GARCH family (both short and long run) with the convoultional neural network were combined and using the combined models, the efficiency of the main stock index and the five selected indicators for the next 10 days were predicted step by step and its accuracy Based on the evaluation criteria.

After the broadcast world and the epidemic of pandemic covid-19 was a severe economic crisis, For this reason, the need for more prediction became apparent. One of these methods is time series prediction. In this study, first, the effect of covid-19 disease on price of Ethereum and Bitcoin, and the results show that this disease had a negative effect on world prices of Ethereum and Bitcoin. In the next step, using univariate time series methods and with the help of ARIMA models, a model for predicting which is the best model AR (1) and MA(1) and time differentiation was designed, the one-year and two-year forecasts were done with the designed model. According to the reports of the World Health Organization, there is probably corona pandamic for up to one year, and For the next two years, Corona has emerged from a pandemic is called the post-corona period. The results show that After a short decline and reacting to resistance and support, they will have an annual upward trend.

The aim of current study was to provide a model for predicting stock price reaction delay based on grounded theory. In the present study,semi-structured interviews have been used as data collection tools and snowball or chain sampling methods and purposeful sampling has been used to select the sample which based on the principle of theoretical adequacy The research data were analyzed using open, axial and selective coding. Results In this study, based on 42 conducted interviews, a total of 607 interview codes, 101 sub-categories (concepts) and 11 main categories were extracted. Then the qualitative model of the research is designed and based on the analysis of data (interviews) the link between the categories in the form of causal conditions, contextual conditions, intervening conditions, strategies and consequences has been conducted. The results indicated that macro factors and market shareholders are effective in predicting the stock price reaction delay.On the other hand, according to these affecting factors, strategies to improve the stock price reaction delay prediction, including the establishment of corporate information and financial statements, corporate information, market performance criteria, management and corporate control which are aroused in the context of affecting factors and interferers, are presented.