The Research Department of Economics and Management of Tadbir Nikan Business, Marketing, and Finance Open 3092-6238 In Press 2027 01 01 Prediction of Capital Market Trends in Volatile Periods through Modeling Investors’ Financial Behavior Based on a Genetic Algorithm Prediction of Capital Market Trends in Volatile Periods through Modeling Investors’ Financial Behavior Based on a Genetic Algorithm 1 17 EN Journal Article 2025 12 24 <p>The present study was conducted with the aim of designing a comprehensive framework for modeling investors’ financial behavior under conditions of volatility in the Iranian capital market and forecasting future market trends using a genetic algorithm. This research adopted a sequential exploratory mixed-methods approach. In the qualitative phase, in-depth interviews were conducted with 20 academic experts and capital market practitioners, and the components of financial behavior were identified using the grounded theory method. In the quantitative phase, a researcher-developed questionnaire was distributed among 350 active investors in the Tehran Stock Exchange during the second half of 2024. Data analysis was performed using structural equation modeling and a genetic algorithm. The results of path analysis indicated that all five research hypotheses were confirmed at a 99% confidence level. Trading strategy, with a path coefficient of β = 0.42, had the strongest effect on investors’ financial behavior, followed by risk-taking (β = 0.34), reaction to volatility (β = 0.28), and liquidity and trading volume (β = 0.19). The most significant finding of the study was the very strong effect of financial behavior on market trend prediction, with a path coefficient of β = 0.68. The structural model fit indices, including GFI = 0.92, CFI = 0.94, NFI = 0.91, and RMSEA = 0.047, were all within acceptable ranges. The coefficient of determination for financial behavior was obtained as 0.68, indicating strong explanatory power of the model. The optimization results using the genetic algorithm showed that prediction accuracy improved from 78.5% to 89.2%, and the mean squared error decreased by 45.2%. The genetic algorithm was able to predict future capital market trends based on investors’ financial behavior with an accuracy of 96%. By developing an integrated and simulation-based framework for modeling investors’ financial behavior, this study demonstrated that the integration of behavioral variables with evolutionary algorithms can significantly enhance the accuracy of capital market trend prediction. The findings confirmed that investors’ financial behavior plays a decisive role in future market developments, and that trading strategy is the most influential factor shaping financial behavior during volatile periods. These results have important practical implications for capital market policymakers, portfolio managers, and investors, suggesting that the combined use of machine learning, evolutionary algorithms, and behavioral data can provide a more efficient framework for risk management and investment decision-making.</p> https://bmfopen.com/index.php/bmfopen/article/download/426/319