Anyone who has watched a stock soar on hype, only to crater weeks later, has witnessed something that classical economic theory has a hard time explaining. Traditional finance models assume investors behave rationally, weighing information coolly and pricing assets based on facts. Yet markets regularly experience bubbles, panics, and viral trading frenzies (think GameStop) that look anything but rational.
A research team set out to investigate why. Their work, published in the Advance Journal of Econometrics and Finance, examines how common mental shortcuts and emotional reactions shape investment choices, and how those choices feed into the strange patterns we see in stock markets. The central takeaway: overconfidence and herd behavior appear to significantly drag down the quality of investment decisions, while anchoring to past prices and fear of losses also play meaningful roles.
The question behind the research
Lead author Muhammad Abdul Rehman of the Department of Commerce at The Islamia University of Bahawalpur, along with colleagues from the Government of Punjab, Erasmus University Rotterdam, and The Islamia University of Bahawalpur, wanted to close a gap in the existing literature. While individual biases have been studied in isolation, less attention has been paid to how multiple biases combine to create market-wide anomalies like asset bubbles or momentum effects (when rising stocks keep rising past what their fundamentals justify).
The researchers focused on four well-known psychological tendencies. Overconfidence is when investors overestimate their own skill or knowledge. Loss aversion describes the human tendency to feel the pain of a loss more sharply than the pleasure of an equivalent gain, a concept developed by psychologists Daniel Kahneman and Amos Tversky in the 1970s. Anchoring happens when people fixate on an initial number (like the price they originally paid for a stock) and struggle to update their thinking when new information arrives. Herd behavior is the tendency to follow the crowd rather than do independent analysis.
Another gap the team wanted to address: most behavioral finance research has drawn on data from what scholars call WEIRD populations (Western, Educated, Industrialized, Rich, Democratic). The team chose to look at investors in emerging markets, including Pakistan, Malaysia, and Indonesia.
How the study was conducted
The researchers used a mixed-methods design, combining survey data with expert interviews. On the quantitative side, they distributed a structured questionnaire to more than 400 retail investors, financial advisors, and fund managers. The survey contained 30 items spread across six constructs, measured on a 5-point Likert scale where participants rated their agreement from “strongly disagree” to “strongly agree.”
Before full deployment, the team ran a pilot with 40 participants to refine the questions. They checked reliability using Cronbach’s alpha (a statistical measure where values above 0.70 suggest the questions consistently measure what they’re supposed to) and ran factor analyses to confirm the survey items grouped as expected.
For the qualitative portion, the team conducted semi-structured interviews with 15 financial professionals. These conversations were transcribed and analyzed using thematic analysis, a technique for identifying recurring patterns in qualitative data, with the help of NVivo software.
The statistical analysis included descriptive statistics, multiple regression (to test how strongly each bias predicted investment decision quality), and moderation analysis (to see whether factors like investor experience changed the relationship between biases and decisions).
What the numbers revealed
The regression model explained roughly 38% of the variation in investment decision quality, which is a reasonably strong result for behavioral research. Overconfidence emerged as the strongest predictor, with a negative beta coefficient of -0.42 and a p-value of 0.045 (below the standard 0.05 cutoff for statistical significance). In plain terms: the more overconfident an investor was, the worse their investment decisions tended to be.
Herd behavior followed closely, with a beta of -0.39 and a p-value of 0.053, just at the edge of conventional significance. Anchoring showed a near-significant negative effect (beta = -0.35, p = 0.062). Loss aversion actually showed a small positive relationship with decision quality (beta = +0.28), though this result did not reach statistical significance. The researchers suggest that a moderate fear of losses might sometimes encourage more cautious, thoughtful choices.
Loss aversion scored highest on the survey overall, with a mean of 4.20 out of 5, suggesting most investors in the sample were quite risk-averse. Overconfidence and anchoring both scored 3.50, indicating moderate prevalence.
The moderation analysis found that investor experience softened the damage caused by overconfidence. More seasoned investors were less likely to let overconfidence distort their choices.
What the experts said
The interviews with financial professionals reinforced the statistical findings. Thirteen of the 15 experts identified emotional investing as a persistent problem. One noted, “Clients panic sell during dips even when the fundamentals are solid.” Eleven mentioned overreaction to news, with another observing that “Most retail investors chase headlines, not value.”
On cognitive shortcuts, ten experts pointed to anchoring on purchase price: “Investors hold because they want the price to return to what they paid.” Nine flagged confirmation bias, the habit of seeking out information that supports what one already believes.
When asked about solutions, 12 of the 15 experts recommended investor education about cognitive biases, 9 supported the use of robo-advisors (automated investment platforms that follow preset rules), and 8 endorsed pre-commitment techniques such as setting fixed asset allocations in advance.
Takeaways for businesses and investors
The research points toward several practical steps. Financial institutions and regulators could expand investor education programs focused specifically on cognitive biases. Robo-advisors, because they follow rule-based logic rather than emotional impulses, may help retail investors avoid panic selling or hype-driven buying. Financial advisors can encourage clients to set automatic rebalancing schedules and predetermined exit strategies so that decisions aren’t made in the heat of market swings.
The researchers also suggest that behavioral finance training could become a standard part of financial planner certification.
A few caveats are worth noting. The sample focused on investors in specific emerging markets, so results may not generalize perfectly to other regions. Several findings (anchoring, loss aversion, herd behavior) did not cross the standard threshold for statistical significance, meaning more research would be needed to confirm them. And the survey relied on self-reported data, which can be influenced by how accurately people assess their own behavior.
