تحلیل مقایسه ای سوگیری های رفتاری مؤثر بر تصمیمات سرمایه گذاران: شواهدی از فراتحلیل پژوهش های تجربی مالی رفتاری (مقاله علمی وزارت علوم)
درجه علمی: نشریه علمی (وزارت علوم)
آرشیو
چکیده
مالی رفتاری از زمانی که پدید آمد توانست پاسخ گوی بسیاری از تناقضات اقتصاد کلاسیک در بررسی واقعیت های اقتصادی در جامعه باشد. مهم ترین نتایج به دست آمده از مالی رفتاری نشان از وجود سوگیری های رفتاری در برخورد با اتفاقات اقتصادی مختلف است. شناخت و بررسی تأثیر گذاری این تورش ها و مقایسه آنها با یکدیگر برای سرمایه گذاران و تصمیم گیرندگان برای جلوگیری از پیامدهای ناشی از این سوگیری ها بسیار مهم و اثرگذار است که مبنای این پژوهش، همین مقایسه بین این سوگیری هاست. در این پژوهش که بر مبنای فراتحلیل پژوهش های تجربی انجام گرفته در حوزه تورش های رفتاری است، 12 تورش اصلی شناسایی شده در کارهای دیگران باتوجه به منابع و کثرت پژوهش های صورت گرفته برای این تورش ها، انتخاب شد. این سوگیری ها براساس 61 مقاله مطالعه شده انتخاب شد، که 24 مقاله داخلی و 37 مقاله از منابع بین المللی بوده اند. داده های استخراج شده از مقاله ها، به نرم افزار فراتحلیل 2[1] وارد شد که طبق نتیجه خروجی از نرم افزار، ناهمگنی در بین نتایج همه پژوهش ها مشاهده شد و به همین دلیل از روش اثرات تصادفی برای تحلیل نتایج فراتحلیل استفاده شد. درنهایت تأثیر این تورش ها بر تصمیم گیری براساس آماره Z و سطح معناداری محاسبه شده توسط نرم افزار تأیید شد و بیشترین اندازه اثر برای سوگیری حسابداری ذهنی به دست آمد.Behavioral Biases in Investor Decision-Making: A Comparative Meta-Analysis of Behavioral Finance Research
Behavioral finance challenges traditional economic theories by demonstrating how cognitive and emotional biases systematically influence investor decisions. This study conducts a meta-analysis of 61 empirical studies (24 domestic, 37 international) to compare the effects of 12 prominent behavioral biases, selected based on their prevalence and diversity in the literature. Employing a random-effects model in CMA2 to account for heterogeneity, we quantify the biases' relative impacts using effect sizes, Z-scores, and hypothesis testing. Results reveal that mental accounting exhibits the strongest effect size, underscoring its dominant role in distorting financial decision-making. Overconfidence, loss aversion, and anchoring also demonstrate significant—though variable—influences. These findings consolidate fragmented behavioral finance research, offering empirical clarity on the comparative weight of key biases in investment behavior. Keywords: Investment, Behavioral Finance, Bias, Capital Market, Meta-Analysis JEL classification: E22، D03، O16 Introduction Prices in financial markets frequently diverge from fundamental values, even with rational participants, challenging traditional theories like the Efficient Market Hypothesis (EMH) and Modern Portfolio Theory (MPT). Classical finance assumes a structured decision-making process comprising problem recognition, solution identification, alternative evaluation, and optimal choice selection, but empirical evidence reveals systematic irrationalities, which behavioral finance addresses by incorporating psychological perspectives to demonstrate how cognitive and emotional biases distort investor behavior. This study synthesizes research on twelve key biases through meta-analysis, confirming their significant impact across markets, where cognitive biases include overconfidence (overestimating knowledge), anchoring (relying on initial reference points), herd behavior (following crowds), representativeness (using stereotypes over data), availability heuristic (overweighting recent information), and mental accounting (categorizing money subjectively), while emotional biases feature loss aversion (fearing losses more than valuing gains), regret aversion (avoiding potential regret), self-attribution (blaming failures on externals), optimism bias (overestimating success), and self-control issues (failing long-term planning). These biases stem from bounded rationality, time constraints, emotions, social pressures, and information asymmetry, with Prospect Theory further explaining irrationalities by showing how investors assess gains and losses asymmetrically, particularly when facing losses. By understanding these biases, markets can develop tools to mitigate their effects and foster more rational decision-making, as recognizing behavioral biases in investment decisions proves crucial for both investors and policymakers to help mitigate irrational choices and avoid unexpected financial risks, while this research aligns with global behavioral finance studies in emphasizing the need for bias-aware strategies to enhance decision-making stability. Methods This meta-analysis synthesizes empirical research on investor behavioral biases through a rigorous four-step methodology: (1) systematic literature review to identify relevant studies, (2) effect size calculation using standardized metrics, (3) heterogeneity testing via Q-statistics and I² to assess consistency, and (4) model selection (fixed- or random-effects) based on heterogeneity levels, with inclusion criteria requiring studies to examine at least one of 12 key biases (e.g., overconfidence, loss aversion), report statistical outcomes (effect sizes, p-values), and cover diverse markets including traditional assets and cryptocurrencies. Heterogeneity testing determined model selection, where studies demonstrating consistency (Q p ≥ 0.05) utilized a fixed-effects model while those showing significant variation (Q p < 0.05) employed a random-effects model, with effect size analysis testing two competing hypotheses: H₀ (bias X has no significant effect on decisions) and H₁ (bias X has a significant effect), where a Z-test p-value > 0.05 supported H₀ while p < 0.05 rejected it, thereby confirming a bias's influence. The study's key contributions include establishing a unified framework for assessing bias impacts across markets, maintaining methodological rigor through adaptive modeling approaches, and enabling cross-market comparisons that distinguish universal versus context-specific effects, while the findings consolidate fragmented behavioral finance research to offer investors and policymakers actionable insights for mitigating bias-driven risks, ultimately quantifying how cognitive and emotional biases shape investment behavior across different environments to provide robust, generalizable conclusions for improving financial decision-making. Conclusion and discussion This meta-analysis confirms that cognitive and emotional biases universally impact investor decision-making across global markets. The study used a random-effects model, justified by confirmed heterogeneity (I² and Q statistics with p<0.05). Results demonstrate all examined biases significantly influence financial behavior. Key findings reveal several important patterns. Overconfidence bias was universally validated. Anchoring bias showed mixed results, with one dissenting study. Herding and representativeness biases received unanimous support. While some studies contested loss aversion, availability, and regret aversion biases, the majority confirmed their significance. Notably, mental accounting emerged as the most influential bias. The research highlights how bias manifestation varies based on personal experiences and environmental factors. This suggests their prominence depends on context. Emotional biases appear particularly susceptible to recent events and temporal factors. Several limitations should be noted. There are insufficient quantitative studies on lesser-known biases and emerging markets. Future research should prioritize four key directions: First, longitudinal studies of bias evolution. Second, examining interaction effects between multiple biases. Third, investigating market-specific manifestations (traditional vs. crypto markets). Fourth, developing evidence-based mitigation strategies. The cryptocurrency boom presents a critical research gap, as most studies focus on traditional markets. Interdisciplinary collaboration with psychologists could yield practical interventions. Additionally, real-time behavioral tracking in digital markets may uncover new bias patterns. These advancements would help investors and policymakers counteract systematic decision-making errors. Such tools are especially valuable in today's increasingly complex financial ecosystems. The consistency of findings across domestic and international studies underscores how fundamental behavioral biases are to financial decision-making processes.
تبلیغات
