Designing a Composite Comprehensive Stress Index for the Tehran Stock Exchange Using a Machine Learning Approach
This study aimed to design and validate a comprehensive index to monitor systemic risk and market-wide stress in the Tehran Stock Exchange using advanced econometric models and machine learning algorithms. Daily time-series data of selected Tehran Stock Exchange indices from 2014 to 2024 were analyzed. Logarithmic returns were calculated, and the DCC-MGARCH model was applied to estimate the dynamic conditional correlation matrix and systemic risk metrics such as ΔCoVaR. To determine feature importance and optimal weighting of indices, three supervised learning algorithms (support vector regression, artificial neural networks, and random forest) were compared, with random forest selected due to superior predictive accuracy. The composite stress index was then constructed and validated using time stability analysis, stress (shock) testing, and logistic regression forecasting. The results revealed that automobile, real estate, paper products, and metal products sectors carried the highest systemic risk, while computer, coal, and textiles showed the lowest. The comprehensive stress index provided reliable early warning signals during market turbulence and achieved strong predictive performance, with an AUC of 0.801 and an accuracy of 89.9% in logistic regression analysis for shock detection. The developed composite stress index is a robust and dynamic tool for identifying vulnerability points and forecasting systemic crises in the Tehran Stock Exchange. It offers significant practical value for policymakers, market analysts, and regulatory authorities to strengthen market resilience and implement proactive risk management strategies. Incorporating macroeconomic variables and extending the historical dataset could further enhance its accuracy and generalizability.
Presenting the Relationship Model between Economic Policy Uncertainty, Unexpected Earnings, and Voluntary Disclosure of Information (Case Study: Companies Listed on the Tehran Stock Exchange)
This study aimed to design and test a model explaining the relationship between economic policy uncertainty, unexpected earnings, and the level of voluntary disclosure among companies listed on the Tehran Stock Exchange. The research was descriptive and ex post facto, relying on historical financial data. The statistical population comprised all listed companies from 2011 to 2019, and 127 firms were selected through systematic elimination. Data were extracted from financial statements and board reports and analyzed using panel regression with F-Limer (Chow) and Hausman tests. Unexpected earnings were measured as the deviation between actual and forecasted earnings per share, while voluntary disclosure was assessed using a six-component disclosure index. Economic policy uncertainty was calculated based on the inflation rate. Results revealed a negative and significant effect of economic policy uncertainty on unexpected earnings, indicating that higher policy uncertainty decreases the likelihood of generating positive unexpected profits. Conversely, the impact of economic policy uncertainty on voluntary disclosure was not significant, showing no direct relationship between policy uncertainty and managers’ disclosure behavior. The determination coefficients were 49% for the unexpected earnings model and 89% for the voluntary disclosure model, confirming acceptable explanatory power. The findings highlight the importance of stable economic policies in improving earnings predictability and reducing investor risk while suggesting that policy uncertainty alone does not significantly drive voluntary disclosure decisions. Policymakers should aim to minimize economic volatility and enhance transparency to strengthen investor confidence and market efficiency.
A Behavioral Finance-Based Model for Pricing Digital Assets (Decentralized Assets)
This study aimed to develop a conceptual model for pricing digital assets by integrating behavioral finance perspectives and identifying psychological and social factors influencing investors’ decision-making in decentralized markets. A qualitative grounded theory approach was adopted. The study involved 15 experts in digital currencies, blockchain, and behavioral finance selected through purposive sampling until theoretical saturation was achieved. Data were collected via semi-structured interviews and textual content analysis. Open, axial, and selective coding were applied to build the theoretical framework. Reliability was confirmed using quality control indices such as Krippendorff’s alpha, Holsti coefficient, Scott’s Pi, and Cohen’s Kappa, all indicating high inter-coder agreement. The resulting model captured multiple determinants of digital asset pricing. Causal factors included emotional and psychological behaviors (e.g., fear of missing out, fear and greed), the influence of news and media, and social association effects. Contextual factors encompassed uncertainty, ambiguity, and market volatility. Strategic factors such as market trust and credibility, investors’ knowledge and awareness, and reference points were identified. Core conditions included regulatory and legal environments, technological infrastructure, and macroeconomic conditions. Consequences involved enhanced market transparency, analysts’ and advisors’ influence, institutional and retail investor interactions, and the impact of past experiences on risk-taking. The proposed behavioral finance-driven model demonstrates that digital asset pricing extends beyond classical economic frameworks, heavily shaped by investor psychology and external information dynamics. The findings can guide investors toward more rational strategies and support policymakers in creating effective regulations and safer decentralized financial ecosystems.
Evaluation of the Model of Factors Influencing Cash Management and Illusion
Cash management and illusion are among the most vital aspects of financial management and play a key role in organizational success. Accordingly, the present study aimed to evaluate the model of factors affecting cash management and investor illusion. This research, in terms of overall strategy, was quantitative; in terms of implementation, it was field-based; and in terms of analytical technique, it was descriptive–correlational. The statistical population of this study included experts and managers of investment companies. The sampling method was convenience sampling. The sample size, based on the Krejcie–Morgan table and considering an error level of α = 0.05, was determined to be 72 individuals. To collect data, a researcher-made questionnaire on cash management and illusion (reliability α = 0.66) was used. The validity of the instrument was confirmed by management and psychology scholars. Data were analyzed using SPSS 22 and Smart PLS software through the structural equation modeling method. The causal conditions, through axial categories, influenced investment strategies and their consequences. In addition, intervening conditions and contextual conditions also had a direct impact on strategies. The research findings regarding the validation of the proposed model, after testing the paths, indicated the confirmation of the paths and the presented model. By achieving a better understanding of the factors and mechanisms influencing cash management and illusion, it is possible to improve financial decision-making and reduce the risks arising from investor illusions.
Testing the Threshold of the Expectation Gap in Declared Versus Assessed Taxes as a Driver of Tax Evasion
This study aimed to examine the role of the gap between declared tax and assessed tax as a driver of tax evasion, with a focus on comparing the impact of internal versus external organizational factors and technical versus standard-related conflicts. The present research was quasi-experimental and applied in nature. Data were collected using a questionnaire administered to 150 experienced members of tax dispute resolution boards, who were purposefully selected and divided into two equal groups. Data analysis was performed using analysis of variance (ANOVA) and t-test methods in SPSS software. The findings confirmed both main hypotheses and demonstrated that the expectation gap is a significant driver of tax evasion, although the intensity of this effect varies across dimensions. Results related to the first hypothesis indicated that the expectation gap in “internal organizational factors” is a stronger driver of tax evasion compared to “external organizational factors.” Specifically, the mean impact of internal organizational factors (62.17) was significantly higher than that of external organizational factors (57.13). In the ranking of factors, “excessive administrative bureaucracy” was identified as the most important internal organizational factor, with an impact percentage of 31.28%. Additionally, the findings for the second hypothesis revealed that the expectation gap in “technical (operational) conflicts” is a stronger driver of tax evasion than “standard-related conflicts.” The mean impact of technical conflicts (48.64) was significantly higher than that of standard-related conflicts (43.29). Within this dimension, “differences in the timing of recording items in accounting compared to tax timing methods” was identified as the most influential technical conflict. In the overall conclusion, it can be asserted that although the gap between declared and assessed tax is inherently challenging, the primary focus for reducing tax evasion should be on removing internal organizational barriers and issues within the tax administration system (such as bureaucracy), as well as resolving technical and operational inconsistencies between accounting standards and tax regulations. These findings suggest that internal and technical reforms within the tax authority are far more effective than purely macro-level policy or cultural interventions in reducing the expectation gap and improving tax compliance.
Proposed Network Model of Management Accountants’ Networking and Strategic Management Accounting Considering the Role of Organizational Culture in Healthcare Organizations
The purpose of this study is to contribute to the limited body of contingency theory literature concerning the determinants of strategic management accounting (SMA) techniques and the role of management accountants in the country’s public healthcare sector. The research, in terms of nature and method, is descriptive and correlational. The statistical population consisted of management accountants and financial managers working in public healthcare organizations. A total of 213 questionnaires were collected. Structural equation modeling (SEM) was employed for the statistical analysis. The findings of this study document a positive relationship between management accountants’ networking and the implementation of SMA techniques (β = 0.367). However, the study found no empirical support for moderating effects of a result-oriented organizational culture (β = −0.138) and an innovation-oriented organizational culture (β = −0.087). Instead, an innovation-oriented organizational culture showed a significant positive indirect effect (β = 0.140) on the implementation of SMA techniques through management accountants’ networking, but not a direct effect (β = −0.062). Conversely, this study demonstrated both direct (β = 0.160) and indirect (β = 0.083) positive effects of a result-oriented organizational culture on the implementation of SMA techniques. Hospital managers and policymakers in healthcare organizations can enhance the application of SMA by paying closer attention to the significance of management accountants’ networking and the characteristics of organizational culture.
Predicting Extreme Losses in Periods of Financial Stress Using the Asymptotic Marginal Expected Shortfall Decision Criterion
This study aims to examine and predict extreme losses during periods of financial stress by applying the asymptotic marginal expected shortfall (AMES) criterion and to compare its predictive capacity with traditional systemic risk measures. The AMES measure was theoretically defined and formulated based on multivariate extreme value theory to model tail dependencies between banks. Daily price data of twelve active banks listed on the Tehran Stock Exchange from 2017 to 2022 were analyzed. Each bank’s contribution to systemic risk was calculated, and the predictive performance of AMES was tested against the conventional marginal expected shortfall. The results revealed that AMES significantly outperforms the traditional marginal expected shortfall in forecasting severe losses during extreme systemic events, offering distinct and richer information about systemic vulnerability and bank-level risk contributions. The additive property of AMES makes it a robust and practical tool for fair systemic risk allocation among banks. Relying solely on simple characteristics such as size or individual risk is insufficient; AMES provides a more accurate basis for supervisory decisions and capital requirement design in times of financial crisis.
Audit Marketing in the Relational Paradigm: Audit Firms in Iraq
This study aims to identify and conceptualize the core elements of audit marketing in Iraqi audit firms using the relational paradigm to provide a localized framework for competitiveness and sustainable client relationships. This exploratory qualitative research employed a grounded theory approach (Strauss and Corbin). Data were collected through a systematic literature review and semi-structured interviews with 15 experts, including senior academics from top Iraqi universities and senior partners of major audit firms. Purposeful snowball sampling was applied, and data were analyzed through open, axial, and selective coding to build a conceptual model of audit marketing. The results revealed that causal conditions driving audit marketing include intense market competition, demand for comprehensive services, the importance of reputation and trust, privatization, digital technologies, and regulatory requirements. Contextual factors encompass regulatory and cultural environments, limited infrastructure, political instability, and workforce knowledge gaps. Intervening conditions such as corruption, economic volatility, threats to professional independence, and resource constraints influence the process. Key strategies identified include service diversification, networking, digital marketing, reputation management, and the development of communication skills, leading to outcomes such as client loyalty and satisfaction, improved audit quality, growth in non-audit revenues, and reduced marketing costs. The findings highlight that success for Iraqi audit firms in complex and unstable environments depends on adaptive marketing strategies, sensitivity to sociocultural contexts, and leveraging relational approaches to foster trust and long-term client engagement. The proposed model offers practical guidance for professional marketing development and organizational resilience in challenging economic and political conditions.
About the Journal
Owner: Research Institute for the Development of Knowledge and Research
Publisher: Maher International Publication
Phone: +982166859278
Address: No. 25, 37th Street, After the Third Roundabout, Tehran Pars, Tehran.
Accounting, Finance and Computational Intelligence is a prestigious open-access journal dedicated to advancing scholarly research at the intersection of accounting, finance, and computational intelligence. The journal provides a dynamic platform for academic researchers, industry professionals, and policy-makers to share cutting-edge developments, empirical studies, theoretical advancements, and applications of computational tools in solving complex problems in accounting and finance. Our commitment to fostering innovation is reflected in the journal's diverse scope, which encourages interdisciplinary research that bridges gaps between finance, accounting practices, and computational intelligence.
We believe that the future of accounting and finance lies in the seamless integration of artificial intelligence (AI), machine learning (ML), and other computational methodologies to enhance the accuracy, efficiency, and predictive power of financial models and decision-making processes. The journal invites submissions that contribute to theoretical advancements, provide practical insights, or present case studies that demonstrate the power of computational intelligence in reshaping the financial landscape.
Current Issue
Articles
-
Identifying the Dimensions and Components of Organizational Governance in Artificial Intelligence-Based Higher Education
Mashallah Salehpour ; Ali Akbar Ramezani * ; Seyyed Hossein NaslMousavi , Mirsaeid Hosseini Shirvani1-13