mahdi madanchi zaj

Digital Economy, relying on emerging technologies like Fintech, is experiencing rapid growth and has resulted in the change of financial and economic models worldwide, thus potentially serving as a crucial factor in Digital Finance development. Considering the speed of these changes, policymakers and industry players should put the digital economy development program and the designing of new models on their agenda. In recent years, Islamic finance has witnessed significant growth due to digital technologies, and many Islamic countries have compiled strategic plans and a clear vision for the Islamic Digital Economy (IDE), which has created fierce competition between them. Therefore, this research aims to design a model of Digital Finance development based on I-Fintech and draw a roadmap for the IDE in Iran. The research method is the Multi-Grounded Theory which is a developed alternative approach to the grounded theory. The statistical population consisted of theoretical and experimental experts, chosen via purposive sampling. The research data was analyzed using MAXQDA and NVIVO in the qualitative section and Smart PLS and AMOS software in the quantitative section. Data analysis led to the identification of 6 paradigms, 36 main categories, and 64 sub-categories, and the research model was designed. The research findings confirm that the strategies and implications of the research model include the Compilation of an IDE Document, Building Regulatory and Supervisory Sandboxes, and establishing Regtech and Suptech. Finally, policy programs and operational solutions for the development of the Digital Finance Ecosystem in Iran are presented.

The design of automated trading systems has increased in various countries, including Iran, recently. The advantage of such systems is the increase in speed and accuracy of trading decisions without emotional interference. Having a model that can simultaneously estimate opening, maximum, minimum, and closing prices is a significant advantage for investors. In this study, using the LSTM neural network, simultaneous estimation of these four prices is performed, and various classification algorithms are used for buying, selling, and holding points to design an optimized algorithmic trading system using a genetic algorithm. To evaluate the proposed model, data from Shapna, Khodro, and Fajr stock markets in Iran from 2012 to 2020 and MATLAB software are used. The results of this research show that the proposed method has a very desirable performance in both simultaneous multi-price estimation and classification of buy, sell, and hold classes. Therefore, it can be considered a very suitable method for automated trading, and a robot based on this method can be designed and implemented in the Iranian stock market for this purpose.

The purpose of this research is to provide a decision support system to assess the credit risk of real bank customers. For this purpose, firstly, using the library method, the effective indicators on the credit risk assessment of real customers were identified and then, using the fuzzy Delphi method, the indicators were screened. The community of this section was formed by banking experts (Bank Mellat) who were selected by the snowball method. In the following, The decision support system was designed and implemented to evaluate the credit risk of real bank customers in SPSS Modeler software. The final data related to the indicators included the files of 7318 real customers of Bank Mellat during the years 2013-2016.The results of the research showed that the decision support system consisting of support vector machine and random forest correctly predicted the status of customers in the four categories of on-time delivery, past due, overdue and questionable delivery by 67.09% and 65.10%, respectively. Meanwhile, the combined support vector machine-random forest model has provided a better prediction with a rate of 77.17%.

The present study, entitled Comparison of artificial neural network method and hidden Markov model in predicting Tehran Stock Exchange index, was classified as applied, analytical-mathematical research, the local territory of those companies listed on the Tehran Stock Exchange and its time domain is from 2007 to 2017 that in terms of data collection, it is a post-event research, in order to analyze information from statistics and mathematics, the Markov model of secret-neural network model has been used. According to the MAPE index, the artificial neural network method has been able to improve the prediction power by 0.0343% compared to hidden Markov's model. Artificial neural networks with the ability to deduce meanings from complex or ambiguous data are used to extract patterns and identify methods that are very complex and difficult for humans and other computer techniques to be aware of. A trained neural network can be considered as an expert in the category of information given to it for analysis. As a result, due to the complexity and heavy calculations, as well as the long computation time and the lack of access of some researchers to advanced models and Markov's secret model is recommended for those who are looking for a simple, fast and reliable method of forecasting using the artificial neural network method to predict the price of stock indices

This study examines the prediction of volatility through jump risk in Exchange Traded Funds (ETFs) on the Tehran Stock Exchange, aiming to play a crucial role in the pricing options of financial assets through volatility forecasting mechanisms.

The study employs three main families of heterogeneous autoregressive (HAR) models for volatility forecasting, incorporating jumps within an econometric framework by estimating heterogeneous autoregressive models. By comparing various models, the study utilizes high-frequency data-based realized volatility to provide precise measurements. The data set includes daily and 15-minute interval data collected from six equity and fixed-income funds over the financial period from 2020 to 2022.

The results indicate that among the funds, the predictive power of volatility is higher in equity funds. Additionally, the findings reveal that the second-order heterogeneous autoregressive model with jumps (HAR-J) is the most effective heterogeneous autoregressive model for modeling and forecasting realized volatility, as determined by the mean squared error and quasi-likelihood criteria. Furthermore, strong evidence supports that second-power variation-based models outperform their counterparts in predicting realized volatility, offering more accurate forecasts and better volatility estimations than second-order heterogeneous autoregressive models.

This study provides insights into the differences among heterogeneous autoregressive models, specifically within the heterogeneous autoregressive model group, so that its results can be used by analysts and fund managers for improved financial decision-making.

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.

Investment decisions are examined based on financial concepts that are based on the efficient market hypothesis. Considering the confirmation of the complete inefficiency of Iran's capital market, there is an expectation of obtaining more returns. In this research, we seek to introduce investment portfolios based on the momentum investment strategy (which has been confirmed in various studies) and its combination with other factors affecting the company's value and profitability. Based on this, 119 companies were investigated in the period of 1388 to 1400 in three short-term periods of three and six months and medium-term of one year. According to the findings, the average returns obtained from the formation of different combined portfolios are positive in all periods and the returns of the combined portfolios are higher than the single-criteria portfolios. The investment strategy based on the asset growth rate and momentum in the short-term periods of three and six months and the medium-term of one year has the highest average return, then the strategy based on over-leverage and under-leverage and momentum and finally the strategy based on leverage

Understanding the potential of cryptocurrencies in the country's economy leads to targeted investment in them and stabilization of the investment portfolio value. Risk hedging of capital through cryptocurrencies can be a novel potential for preserving the value of the investment portfolio. In this study, the factors affecting the risk hedging of conventional investment portfolios using cryptocurrencies and the nature of the relationships between these factors have been identified.

In the present study, first, by examining domestic references, common investment methods forming the investment portfolio that are of interest to the general public in the country were identified. Then, by conducting an integrated review of existing references in the field of cryptocurrencies and other conventional investment methods, the factors influencing the hedging capacity of the investment portfolio using cryptocurrencies were identified. Subsequently, utilizing the opinions of experts including university professors and capital market managers, the relationships between variables were determined. Based on the interpretive structural modeling approach, a model of factors affecting the risk hedging of common investments in the country using cryptocurrencies has been presented.

The results of this study showed that the most influential factors in using cryptocurrencies for investment risk hedging in Iran are international political factors and global prices of gold and oil. On the other hand, the most dependent variables are related to the lack of user knowledge and regulations in the field of cryptocurrencies within the country.

Originality/Scientific Value-Added:

Based on the conducted research, a model is proposed to identify factors affecting capital risk hedging in the country with the help of cryptocurrencies. Considering the obtained relationships, it was determined that factors affecting the understanding and application of cryptocurrencies in Iran do not have an impact on the risk of conventional investment methods in Iran. Furthermore, in hedging investment risks in Iran using cryptocurrencies, international factors are considered the most influential factors.

There has been an increase in the development of automated trading systems in numerous countries, including Iran. The greatest advantage of such systems is that they allow traders to make trading decisions at an increased pace and with more accuracy without having to rely on emotions. Estimating price is one of the most important aspects of algorithmic trading. Deep neural networks are preferred for estimation. Additionally, investors who rely on algorithmic trading have a huge advantage by having a model that estimates opening, maximum, minimum, and closing prices simultaneously. In this study, using short-term LSTM deep-long-term memory neural networks, and the genetic algorithm, these four prices are estimated simultaneously. In addition, the optimal feature was selected by considering 40 price, volume, volumetric and volumetric indicators. The proposed model is evaluated using five shares from the Tehran stock exchange during the period 2012-2021, namely Isfahan oil refining, Iran Khodro, and Amirkabir Kashan Steel, Eqtesad Novin Bank, Chin-Chin Industry and Cultivation, and Exir Pharmacy. Based on the results of this study, the proposed model has excellent simultaneous estimation performance and the average estimation error of all 4 prices is less than 8%, demonstrating that the proposed method has a lower estimation error.

AbstractThis research tries to determine and prioritize the effective factors on improving the performance of capital market specialists in the Tehran Stock Exchange by studying market timing and strategies for investors to make correct and responsible decisions regarding their capital. The method of conducting this research is based on the results, practical; based on Objective, descriptive; Based on the amount of researcher control, non-experimental; based on data type, quantitative; And based on the research method, it is retrospective. The sampling method was a targeted judgment method from among experts in the capital market, and finally 15 people were selected. In this research, the test of the normality of the distribution was done by using the absolute value of the skewness and kurtosis of the variables, and the results indicated the normality of the data. Bartlett's test (K M O) was used to check the adequacy of the sample size. In comparing the market timing strategy models, the results showed that the season and date timing strategy, compared to the strategies of using the moving average and Sy Harding's seasonal timing, is more effective on the managers' performance and predicting the future movements of the market; But compared to Mark Walker's seasonal timing strategies and Hirsch's best six months timing (B.S.M), it has less efficiency.

Due to the high market share and the leadership of Bitcoin’s price in the cryptocurrency market, this crypto asset has always been the leader of the trading trends of digital currencies and its movements have a significant impact on the price trends of other cryptocurrencies. In this article, by using a combination of meta-heuristic and machine learning methods, a model with the least error for estimating the price of Bitcoin is presented. This research uses the stacking approach of popular machine learning algorithms along with optimizing the parameters of the algorithms. In this study, for the first time in the field of cryptocurrencies, the differential evolution method has been used to find the most optimal stack combination and also to identify the most suitable parameters of each algorithm. This research in terms of type, data collection method and purpose is descriptive, modeling, and applied ,respectively.In this study, the model results have been compared in three scenarios with inputs including OHLC features, technical indicators (more than 160 indicators), fundamental analysis indices (7 indicators) and their combination. Considering that most articles have applied technical indicators as input to machine learning models, the use of fundamental indices is one of the distinguishing features of this research. The results show that depending on the features used in each scenario, the type and order of placement of learning algorithms in the price prediction stack changes. All three scenarios investigated in this research have acceptable accuracy for price prediction.

The main goal the model and optimal investment portfolio selection to maximize stock portfolio returns based on the forecasted price and minimize investment portfolio risk based on the Markowitz model. This paper presents is to select the optimal stock portfolio based on data training through Markov decision-making and ensemble learning. To teach data from the data of five years (2016-2011), 85 active stock exchange companies in Iran that have been filtered based on technical, fundamental, and time series variables have been used. Therefore, the stock sets are first filtered based on optimizing trading rules based on technical analysis, Markov decision-making and ensemble learning that issued the buy signal. Data for the next 5 years (2020-2016) were also used to test NSGA II and MOPSO algorithms. According to the obtained results, if the shares are bought equally among 85 companies and maintained for five years, the average return on the total stock portfolio is equal to 13.08%, with a risk of 0.946%. While using the MOPSO algorithm has achieved an average of 43.54% with an average risk of 1.102% . The rate of return on capital for the NSGA II algorithm was also the highest in 5 years. Therefore, it can be said that based on the obtained indicators, NSGA II algorithm is the best combination of the stock portfolio.

Forecasting the stock price index volatility is considered a strategic and challenging issue in the stock markets, and it is momentous for traders and investors in the decision-making process. Hence, the presentation of an efficient model for forecasting the stock price index volatility is a crucial and hard task because stock market data and price fluctuations have high volatility and nonlinearity characteristics. To beat this challenge, this paper proposes a new hybrid model by applying artificial intelligence algorithms to forecast the stock price index. It incorporates four phases to provide a dynamic and exact model: (1) Select popular and key technical indicators as input variables (2) Apply Adaptive Neuro-Fuzzy Inference System (ANFIS) for designing a substructure to provide a high-quality and quick solution (3) Use Modified Particle Swarm Optimization (MPSO) to enhance predictive accuracy by simultaneously and adjusting the length of each interval in the discourse universe and the appropriate degree of membership (4) Employ Parallel Genetic Algorithm (PGA) to solve complex issues with computational weight optimization and adjusting the decision vectors employing genetic operators. The stock market data of “Tehran Stock Exchange (TSE)” from 01/01/2011 to 31/12/2021 are utilized to examine the functionality of the proposed model. In comparative assessments, the overall performance of the ANFIS-MPSO-PGA model based on 5 criteria achieved 81.45%, which was significantly better than other methods.

Traditional models do not have sufficient ability to predict the return on investor portfolio due to changes in the external environment (systematic risk) and the internal environment (non-systematic) and this is mainly due to the identification of the explanatory variables and the experimental design of the model.Therefore, the present research, while explaining this issue and in order to adjust the uncertainty problem of the model, by averaging all the models (Bayesian averaging), has determined the effective risks on stock returns in Iran.The present study expresses this failure in identifying explanatory variables and empirical model design. The statistical sample of the research includes 138 listed companies in the period 1390 to 1399.In this study, 62 risks affecting stock returns in the form of 31 indicators in the field of systematic risk and 31 non-systematic indicators entered into nonlinear Bayesian models with time-varying parameters.The results show that among BMA, TVP-DMA, TVP-DMS, BVAR and OLS models, the TVP-DMA model is the most efficient model. According to the TVP-DMA model, 10 non-fragile risks include systematic risks (real GDP growth rate, unofficial market currency, inflation rate, interest rate) non-systematic risks (instantaneous ratio, liabilities, cash flows from operations, return on equity, debt ratio, and price-to-earnings ratio) as the most important risks affecting stock returns. All the mentioned risks, except interest rate and debt ratio, have a positive effect on stock returns.

One of the most important challenges for Islamic financial engineers is to pay attention to the differences between religions in Islamic countries, including Iran, and plan to discover the differences, resolve legal ambiguities, and finally lay the groundwork for increasing public participation in various financing projects. The importance of the present research is to resolve the legal ambiguities and doubts in order to increase participation in the financing of Mudarabah bonds. The lack of financial research in Iranian Sunni jurisprudence in connection with the issue of Mudarabah bonds, the population and the high economic and development potentials of Iranian Sunnis, as well as the potential ability to participate in various financing projects by this special segment of the Iranian society, from It is one of the necessities of conducting the present research. . Therefore, in this research, with the aim of reducing the jurisprudential doubts of the Iranian Sunnis, from the two perspectives of Shafi'i and Imami jurisprudence, the Mudarabah papers were examined and analyzed jurisprudently, in order to clarify and resolve its jurisprudential ambiguities, in fact, the present study seeks to answer this question. What are the similarities, differences and the percentage of jurisprudence and difference in Mudarabah papers based on the standards of Imamiyyah and Shafiiyyah jurisprudence. The main approach in this research is a comparative approach. In order to analyze and issue jurisprudence in Imami jurisprudence, the document analysis method (documents and approvals of the Jurisprudence Committee of the Stock Exchange Organization) has been used, and in connection with Sunni jurisprudence (Shafi'i jurisprudence), in-depth interview and focus group methods have been used, a statistical sample It includes 9 experts of Shafi'i jurisprudence. The results of the research show that the similarities and common points of jurisprudence are very high and the percentage of jurisprudence and jurisprudence in the two jurisprudence systems of Imamiyyah and Shafiiyya in relation to Mudarabah papers is at a very high level (100%), which is important. It can be the basis for increasing the participation of Iranian Sunnis in projects related to Mudarabah bonds and other Islamic securities issued in Iran.

The importance of predicting and knowing the future in order to plan and formulate economic strategies is not hidden from anyone. The accuracy of forecasts is one of the most important factors in choosing the type of forecasting method. The stock price index is one of the effective variables in economic systems that these very complex time series are usually assumed to be random and as a result their changes are assumed to be unpredictable. Such time series variables have the property that the shock to the variable takes a long time to disappear due to the possibility of long-term memory. The aim of the present study was to test the fractal market hypothesis with the Markov regime change model with a possible combination and convergence in the Tehran Stock Exchange. In this paper, the amount of long-term memory and stability of financial time series resulting from the total stock market index for the period 1388-1386 were examined. For this purpose, first the existence of long-term memory was examined, then the fractal nature of the market was examined using the Harst view index. The results indicate the existence of long-term memory in this variable. In this case, with one differentiation, it becomes more differentiated, so the stock price index series in Iran has long-term memory and the effects of each shock on this variable due to its long-term memory remain for long periods. It stays. The results also showed that the overall stock market index is fractal.

In this study, we provide a model for evaluating the performance of capital market professionals (Case Study of Mutual Fund Managers) on how to apply market timing strategy in predicting future market movements, using a combination of multi-indicator decision making and We have analyzed the scenario of "expert system proposed by Hosseini Nasab" using the network analysis process (ANP) [3]. In the mentioned method for modeling and analyzing data, the matrix of internal and external dependencies of strategies and criteria was calculated and using the designed network model, strategies for determining weight and the most important strategies were evaluated. This method is used to implement the model of capital allocation to corporate stocks, with the three objectives of return, risk and liquidity.

Due to the importance of predicting economic variables, different models have been created to predict the future values of variables. In fact, economic models can be tested by checking the level of forecasting accuracy. The main purpose of this study is prediction of Iran interbank offered rate and Iran treasury exchange rate as interest rates indicators for facilitating interest rate risk management. Two econometric models including ARFIMA and ARIMA have been used for forecasting. Thus, the ARFIMA model considering long-term memory and the ARIMA model without considering long-term memory have been considered. The evaluation of the prediction accuracy of the two models using the monthly Iran interbank offered rates data and also the monthly Iran treasury exchange rates data shows that both the interbank offered rates data and the Islamic treasury bond rates data, ARIMA model has a better performance compared to ARFIMA model in predicting data.

AbstractThe purpose of this study is to study the market timing and provide a model for investors to make the right and responsible decisions against their capital and tries to examine the capabilities of man-angers of mutual funds in the Iranian capital market, whether they are time skills Have market assessments in choosing their investment portfolio or not. Therefore, the variables of the present study, using the implementation of a combined method of multi-attribute decision making and scenario analysis of "expert system proposed by Hosseini Nasab" [2], we used the A N P method. The results showed that it is possible to provide a comprehensive "timing" model for the capital market and In-prove the performance of professionals, and the future movements of the market can be predicted using the market timing strategy; Also, comparing the models of market timing strategy, the results showed that the timing strategy of seasons and dates, in comparison with the strategies of using moving average and seasonal timing of Sai Harding, have an effect on managers' performance and predicting future market movements; But compared to Mark Walker's seasonal timing strategies and Hirsch's best six-month timing strategies, it is less effective.

Application of technical analysis and Markov filterIn future optimal portfolio managementEmphasizing the amount of investor riskMetaphysical artificial intelligence approachAbstractIn decision making for investment, two factors are very important and are the basis of investment. These two factors are risk and return, and in this regard, the study and study of investors to select the best investment portfolio is done according to the amount of risk and its return. A portfolio is a combination of assets formed by an investor to invest. The process of selecting a stock portfolio is one of the issues that has been the focus of many researchers. The aim of this study is to create an optimal stock portfolio using the predicted data. The statistical sample of the research includes the financial data of Iranian stock exchange companies during the years 1390 to 1397. In this study, using stock regression algorithm to predict stock returns, and finally using Markov method and spectral clustering algorithm, the necessary filter to select the appropriate initial data was performed and cultural meta-processing method with prediction data, It provided the optimal portfolio of stocks for the investor group with risk-taking as well as risk-averse. The research results show that the cultural transcendental algorithm, according to Sharp's method, has the ability to create an optimal stock portfolio using predicted data using the Marquis method for venture capitalists and risk averse investors.Keywords: Optimal stock portfolio, cultural meta-innovation algorithm, spectral clustering algorithm, Markov method, technical analysis

 The purpose of this research is to create a model of ethical decision-making that applies to accountants and the accounting profession. In this research, an integrated pattern of five factors affecting ethical decision-making by accountants was identified, which include professional codes of conduct, philosophical orientation, religious orientation, values derived from culture, and moral maturity.

 The present study is a review study in which the articles indexed in scientific databases such as Academic Jihad Scientific Information Center, Marja Danesh, Noor specialized magazines database and financial analysis publication were made. Research texts from 1992 to the end of 2023 have been examined. In this study, the researchers searched and extracted the valid English articles from the valid electronic sources by using the determined keywords, and by examining the full texts of these articles, they described the obtained data in a classified manner.

 According to the findings of the research, most of the financial crises are caused by the non-observance of ethics in decision-making processes, and following ethical principles in decisions and professional behavior can prevent society from moving towards financial crises in the long term.

Originality / Value:

 Based on the findings of the research, it was finally determined that the ethical requirements in investment is one of the categories of the model of ethics in investing in the financial crisis. The organization and development of the capital market depends on the regulation of the structure of the financial market in accordance with the macroeconomic characteristics and the coordinates of the companies active in the country's economy.

The formation and emergence of new scientific branches in management shows the complexity of management. The present study aims to provide a model of investment ethics with financial crises based on the grounded theory approach and this research is applied in terms of purpose and has been done with an exploratory approach and quantitative research design. Participants in the interview consisted of 9 experts and experts in the field of finance and the total number of questionnaires distributed was 180. Interviews were analyzed using theme analysis method using MaxQDA software and statistical analysis was performed using SPSS software version 21. The results showed that 5 main categories, 31 sub-categories and 74 open codes were obtained from the interview. According to the confirmation of the fit of the studied model in Smart PLS software and the general fit indices of GOF and Q2 criterion, the optimal model of investment ethics in the financial crisis was presented.

Interest rate risk is one of the most important financial risks that economic enterprises faced. This risk is actually the probability of a decline in the value from unexpected fluctuations in interest rates in the market. Firms use different analytical models to evaluate the interest rate assessment and the effect of interest rate fluctuations on liabilities and assets and their cash flows. One of the risk hedging strategies is risk management using operational techniques.In this research, with the aim of proposing an operational technique for hedging interest rate risk, the sensitivity coefficient of firms' stock returns to interest rate fluctuations in the period of 2011 to 2021 and after entering the debt market has been investigated and analyzed. In order to measure the interest rate sensitivity coefficient, a model similar to the model of Flannery and James (1984) and Deleze and Korkeamaki (2018) has been used. In this research, the autoregressive integrated moving average (ARIMA)model has been used to estimate the unexpected part of changes in interest rates, and the rolling window regression and panel data models have been used to estimate the interest rate sensitivity coefficient and analyze and investigate this sensitivity coefficient. The result of the research shows a decrease in the sensitivity coefficient to interest rate changes after the first entry into the debt market and finally the facilitation of interest rate risk management among debt bond issuers.

The main purpose of this study is to identify and refine the factors affecting the efficiency of investment in insurance companies based on the fuzzy Delphi approach. In order to identify the factors affecting the financial stability of companies, while reviewing the studies, semi-structured interviews were conducted using a qualitative method of theme analysis. The experts interviewed are 20 university professors, CEOs of insurance companies and institutions, as well as senior auditors with accounting and management degrees. A review of the literature, research, and interview results reveals two main themes, including inputs and outputs, that categorize the factors affecting the efficiency of investment in companies and institutions. In this study, key factors have been identified and then the fuzzy Delphi technique has been used to rank and find the degree of importance of the factors. The results showed that return on assets, return on investment, value of investment, return on investment and net return on capital from the category of outputs and market share variables, ability to fulfill obligations, business diversity, technical reserves and risk transfer from the category of inputs, respectively. Have the greatest impact on the investment efficiency of insurance companies.

The purpose of this study is to validate the real customers of Bank Mellat using fuzzy Delphi approach and multiple logistic regression. For this purpose, first, the effective indicators on assessing the credit risk of real customers were identified by library method and then, the fuzzy Delphi method was used and the indicators were screened. The community of this section was formed by banking experts (Bank Mellat) who were selected by snowball method. Then, the final data related to the indicators, including the files of 7318 real customers of Bank Mellat during the years 2014-2020, were collected and analyzed by multiple logistic regression method in four categories of timely receipt, past due date, delinquent and doubtful receipt. The results showed that the loan amount, loan repayment time, installment interval, installment amount, loan extension, previous loan, real estate collateral, inventory average, facility interest rate and education level have a significant effect on credit risk of real customers. Also, the significance of gender, age and occupation indices was not confirmed.