Abstract
Purpose
The study attempts to investigate the relationship between emotional biases (loss aversion bias, overconfidence bias, and regret aversion bias) and investment decisions through a meta-analysis approach.
Design/methodology/approach
A meta-correlation analysis was done using sample size and correlation (r) data from several relevant studies that look at how emotional biases (loss aversion bias, regret aversion bias, and overconfidence bias) affect investment decisions. Additionally, beta coefficients (ß) were also converted to correlation coefficients (r) from six studies.
Findings
This study analysed 31 empirical studies and found a significant positive correlation between emotional biases and investment decisions [loss aversion bias (r = 0.492), regret aversion bias (r = 0.401), and overconfidence bias (r = 0.346)]. We set the statistical significance threshold at 0.05.
Research limitations/implications
The review covered 31 online research publications that showed significant heterogeneity, possibly influenced by various methodological, population, or other factors. Furthermore, the use of correlational data restricts the ability to establish causation.
Originality/value
This is a novel attempt to integrate the results of various studies through meta-analysis on the relation between these emotional biases (loss aversion, overconfidence, and regret aversion) and investment decisions.
Keywords
Citation
Kumar, S. and Chaurasia, A. (2024), "The relationship between emotional biases and investment decisions: a meta-analysis", IIMT Journal of Management, Vol. 1 No. 2, pp. 171-185. https://doi.org/10.1108/IIMTJM-03-2024-0034
Publisher
:Emerald Publishing Limited
Copyright © 2024, Shailendra Kumar and Akash Chaurasia
License
Published in IIMT Journal of Management. Published by Emerald Publishing Limited. This article is published under the Creative Commons Attribution (CC BY 4.0) licence. Anyone may reproduce, distribute, translate and create derivative works of this article (for both commercial and non-commercial purposes), subject to full attribution to the original publication and authors. The full terms of this licence may be seen at http://creativecommons.org/licences/by/4.0/legalcode
1. Introduction
The study of financial decision-making took a big turn in the 1970s, when “Amos Tversky” and “Daniel Kahneman” first established the notion of behavioural biases (Kahneman and Tversky, 1979; Tversky and Kahneman, 1974). Human cognition and emotions drive these biases, offering fresh insights into investor behaviour and the impact of psychology on financial decisions (Raheja and Dhiman, 2020; Sapkota, 2023; Tversky and Kahneman, 1974). According to Pompian (2012), behavioural biases are categorised into two primary groups, i.e. emotional and cognitive biases.
Emotional biases and cognitive biases are two crucial components of financial decision-making (Pompian, 2012; Ritika and Kishor, 2022; Sapkota, 2023; Thevaruban, 2022). Cognitive biases are related to information processing and errors while making decisions, whereas emotional biases refer to those biases that are caused by emotional factors such as fear, regret, and arrogance.
Emotions act as powerful drivers of human reasoning, whether it is in financial markets or other aspects of our lives (Cotruş et al., 2012). Existing research has looked into how various emotional biases, such as loss aversion, regret aversion, and overconfidence, might influence investment behaviour. However, an exhaustive review of these data to determine the total impact of these biases on investment decisions currently remains missing.
The purpose of this meta-analysis is to bridge this gap by thoroughly examining and analysing prior research on the subject. The fundamental question motivating this research is: How do emotional biases like loss aversion bias, regret aversion bias, and overconfidence bias affect investing decisions?
In this meta-analytical study, we have examined the relation between three prominent emotional biases with investment decisions that are overconfidence bias (investors used to assume that they are best at understanding and analysing market conditions), loss aversion bias (investors saving themselves from losses, and consider losses more important than gains), and regret aversion bias, in which investors make illogical decisions to avoid the fear of regret (Gyawali and Neupane, 2021; Laungratanamas and Nuangjamnong, 2022; Pompian, 2012; Ritika and Kishor, 2022). Through the meta-analysis approach, this study aims to provide a better understanding of these emotional biases in investment decision-making. We will apply a systematic search strategy to discover relevant studies and utilise correlation coefficients to estimate the impact sizes for each emotional bias. The anticipated outputs of this study will be a summary of impact sizes and an investigation into potential variables that may modify these relationships.
2. Literature review and hypothesis development
Ritika and Kishor (2022) classify emotional biases as second-order latent constructs that amalgamate a range of first-order biases. In 2021, Baker et al. identified regret aversion bias and loss aversion bias as emotional biases (Baker et al., 2021). Other researchers have also called them “endowment bias,” “status quo bias,” “self-control bias,” “loss aversion bias,” “overconfidence bias,” and “affinity bias.” However, this research will focus on three prominent emotional biases: regret, loss aversion, and overconfidence (Gyawali and Neupane, 2021).
2.1 Loss-aversion bias
Prospect theory developed the concept of loss aversion bias, which is one of the primary emotional biases that profoundly influence investment decisions (Gupta and Shrivastava, 2022; Kahneman and Tversky, 1979; Pompian, 2012). People often feel the pain of losses more intensely than the pleasure of profits, even when the profits are of the same amount (Kahneman and Tversky, 1979; Kumar et al., 2018). However, research on the topic reveals a surprising inconsistency in findings. Some studies demonstrate a strong influence of loss aversion on investment decisions (Gupta and Shrivastava, 2022; Hunguru et al., 2020; Iram et al., 2021; Kumar et al., 2018), while others report a weaker or even insignificant relationship (Armansyah, 2021; Athur, 2014; Aydin, 2023). For example, Kumar et al. (2018), dealing with the loss aversion tendency of individuals in the Indian stock market, found a significant interaction, while Bhatia et al. (2022), which also examined the usage of robo-advisory services by Indian investors, showed a statistically insignificant relationship. Hunguru et al. (2020) conducted a study on individual investors in Zimbabwe, revealing a robust correlation, in contrast to Athur’s (2014) findings on investors in the Nairobi Stock Exchange, Kenya.
This disparity in studies highlights the need for a thorough meta-analysis to synthesise these data and reveal patterns and trends for the significance and strength of loss aversion bias. Such an approach could provide a more detailed knowledge of its function in investment decision-making, which would enable more focused actions to reduce the effects (Aydin, 2023; Bhatia et al., 2022; Hunguru et al., 2020; Shah et al., 2018).
The emotional bias of loss aversion causes individual investors to prioritise capital preservation over growth because the pain of losses is greater than that of profits (Pompian, 2012; Sapkota, 2023). This tendency is responsible for actions such as holding onto declining investments in the hopes of “winning back” or selling stocks that have appreciated just because of the fear of losing the gains made on them (Pompian, 2012). As a result, if portfolios are already under-diversified, further fluctuations in markets may cause them to depart from their long-term goals and reduce their overall return (Gupta and Shrivastava, 2022; Gyawali and Neupane, 2021).
Based on these observations, we’ve hypothesised that:
There is a relationship between loss-aversion bias and investment decisions (HA1).
2.2 Regret aversion bias
Prospect theory also gives rise to regret aversion bias, which, along with loss aversion bias, is one of the primary emotional biases that profoundly influence investment decisions (Kahneman and Tversky, 1979; Pompian, 2012). Investors who have experienced heavy losses in the past tend to become overly conservative in their investment approach and hence avoid bold or risky decisions (Bhatia et al., 2022; Hunguru et al., 2020; Kahneman and Tversky, 1979). This can result in low long-term capital growth, which can easily harm their investment goals (Octavia et al., 2022).
Similar to loss aversion bias, regret aversion bias also exhibits a remarkable inconsistency across different studies. For example, Iram et al. (2021) conducted a study on women in Punjab, Pakistan, and found that regret aversion bias has a significant relationship with financial literacy but an insignificant influence on investment decisions. Athur (2014) and Elhussein and Abdelgadir (2020), who researched individual investors on the Nairobi Stock Exchange and Khartoum Stock Exchange, also reported the impact as insignificant. In contrast, studies by Chadha (2024), Elhussein and Abdelgadir (2020), Nkukpornu et al. (2020), and Wangzhou et al. (2021) discovered a significant relationship. This variance underscores the complexity and contextual dependence of regret aversion bias. A comprehensive meta-analysis is essential to systematically analyse and synthesise findings across studies, elucidating the significance and intensity of regret aversion bias in investment decision-making. Such an analysis could provide valuable insights into the underlying factors influencing regret-aversion bias, as well as inform future research and interventions aimed at mitigating its effects. Recognising and managing regret aversion bias is crucial for making well-informed investment decisions and achieving long-term financial success (Elhussein and Abdelgadir, 2020; Octavia et al., 2022; Wangzhou et al., 2021).
Based on these observations, we’ve hypothesised that:
There is a relationship between regret-aversion bias and investment decisions (HA2).
2.3 Overconfidence bias
Overconfidence bias is also considered one of the emotional biases that have a profound impact on investment decisions, leading investors to overestimate their abilities and judgement in analysing data or evaluating company reports for investment (Gyawali and Neupane, 2021; Pompian, 2012; Ritika and Kishor, 2022; Sapkota, 2023). Because of their overestimation, they may ignore negative information that should be a potential warning sign, which can lead them to ill-advised stock purchases or reluctance to sell those they already own (Bhatia et al., 2022; Lambert et al., 2012). Moreover, overconfident investors may also engage in excessive trading based on their belief that they possess unique insights that others don’t, resulting in subpar long-term returns (Ahmad and Shah, 2022; Kasoga, 2021; Rahman and Gan, 2020).
Similar to loss aversion bias and regret aversion bias, research on overconfidence bias also shows that there is a considerable difference in the test results about the intensity of bias and involved demographic groups, which vary across situations. For instance, researchers Dhungana et al. (2022), Elhussein and Abdelgadir (2020), and Laungratanamas and Nuangjamnong (2022) identified overconfidence bias among individual investors in Khartoum and Nepal Stock Exchange. Additionally, Adil et al. (2022) highlighted the positive and significant impact of this bias on the investment decisions of male investors in the Delhi-NCR area but found its effect on female investors to be insignificant. Baker et al. (2019) also found the impact insignificant when they examined the Indian stock market. This inconsistency highlights the nuanced nature of overconfidence bias and its manifestation across various demographic groups and contexts.
To gain a deeper understanding of the prevalence and intensity of overconfidence bias in investment decision-making, a comprehensive meta-analysis is essential. Ultimately, this would inform more effective interventions and strategies to mitigate its impact on investment decisions (Ahmad and Shah, 2022; Michael and Oshoma, 2022; Hassan Metwally, 2023; Lambert et al., 2012; Michael, 2023). Based on these observations, we have hypothesised that:
There is a relationship between overconfidence bias and investment decisions.
3. Methods
3.1 Search strategy
To investigate the relationship between emotional biases and investment decisions, an extensive literature search was conducted using a variety of approaches. We searched electronic databases like Web of Science and Emerald using specific keywords like “loss avers*,” “regret avers*,” “overconfidence,” “emotional biases,” and “investment decision*” to find relevant studies. To ensure inclusivity and reduce publication bias and geographical bias, we have also conducted a manual search on Google Scholar, ResearchGate, and Academia. Lastly, we also requested a copy from authors whose works were not available to us through ResearchGate.
3.2 Criteria for inclusion and exclusion
In our initial search, we found 169 studies from Web of Science, Emerald, and 127 studies from other sources. To narrow down our search, we established a set of criteria to refine our results.
All qualitative, viewpoint, and systematic literature reviews have been excluded.
We only analysed those studies that included either loss aversion, regret aversion, or overconfidence and investment decisions. This approach allowed us to ensure that the studies we reviewed directly addressed our research question.
Additionally, studies without data for a meta-correlation analysis were excluded.
After applying these criteria, we left with 31 studies. Each of these studies was analysed and described in Table 1. Based on these criteria, we ensured that only the most relevant and reliable studies were included in this review.
3.3 Effect size used
We require correlation coefficients (r) and sample sizes (n) to analyse the meta-correlation between the variables. Out of the total studies analysed, 25 studies reported r, while 6 studies only reported beta coefficients (β). We have utilised the equation provided by Peterson, R. A., and Brown, S. P. in their study “On the use of beta coefficients in meta-analysis” to compute the correlation coefficient (r) from beta coefficients (β) (Peterson and Brown, 2005). The following equation can be used to compute the correlation coefficient (r):
When the beta coefficient is positive in this equation, then λ = 1, and when it is negative, λ = 0.
3.4 Synthesis method
We performed a meta-analysis of research that looked at the correlation between emotional biases and investment decisions. For data synthesis, the following strategies were used:
Statistical analysis: Jamovi software was employed to calculate meta-correlations between the variables. Weighted effect sizes were calculated for each research project based on the correlation coefficient and sample size. To evaluate the total effect size and its confidence interval, we employed random-effect models with a 95% confidence level. (Hansen et al., 2022; Viechtbauer, 2010).
Heterogeneity Statistics: To evaluate the heterogeneity among the studies, two methods were used.
The Cochran Q test (Cochran, 1954) tests the null hypothesis that all research studies are seeing the same thing, and the adoption of the random effects model in meta-analysis is supported by a substantial p-value.
I2 statistic is used to determine how much variance is due to heterogeneity in studies rather than chance (Borenstein et al., 2009).
Reporting of results: Forest plots and other tabular representations were used in reporting the meta-analysis results.
Publication bias is assessed by using the “Fail-Safe N” and the “Begg and Mazumdar Rank Correlation” tests.
4. Results
4.1 Selected study characteristics
To examine the correlation between these four variables—loss aversion bias, overconfidence bias, regret aversion bias, and investment decision—across different studies, we have extracted the following information: sample size, country, correlation coefficients, author, and year, and also given a code to each study. This information is presented in Table 1.
4.2 Meta-correlation analysis
4.2.1 Hypotheses testing
Table 2 provides the results of a meta-analysis of different emotional biases: regret aversion bias, overconfidence bias, and loss aversion bias on investment decisions. The meta-analysis was also conducted by removing an outlier from the loss-aversion bias data set.
4.2.1.1 Loss aversion bias and investment decisions
The meta-analysis includes 12 studies examining the correlation between loss aversion and investment decision-making. The overall estimate of the correlation is 0.492 with a standard error (S.E.) of 0.198. The Z-statistic is 2.48, and the p-value is 0.013, also indicating a significant relationship between them. However, the analysis reveals high heterogeneity among the studies, with an I2 value of 0.9917 (Table 3). The Q-test with a value of 2424.159 and p-value <0.001 also confirms significant heterogeneity and suggests using a random effect model. Removing the outlier from the dataset resulted in a reduction in the estimate (0.312), but with a lower standard error (0.0848). The Z-statistic of 3.68 and p-value <0.001 indicate a highly significant relationship even after the outlier’s removal. The implication is that loss aversion bias is indeed related to investment decisions, but the effect sizes vary widely across studies, possibly due to different methodologies, sample characteristics, or contextual factors.
4.2.1.2 Regret aversion bias and investment decisions
The analysis includes 14 studies investigating the correlation between regret-aversion bias and investment decision-making. The overall estimate is 0.401 with an S.E. of 0.134. The Z-statistic is 3, and the p-value is 0.003, indicating a significant relationship. Similar to loss aversion bias, there is substantial heterogeneity among the studies, with an I2 value of 0.9861, and a Q-test value of 463.639 (p < 0.001), suggesting significant heterogeneity. This means that while regret aversion bias is related to investment decisions, the strength of this relationship varies significantly across different studies. Removing the outlier from the dataset resulted in a reduction in the estimate (0.285), but again with a lower standard error (0.0576). The Z-statistic of 4.96 and p-value <0.001 indicate a highly significant relationship even after the outlier’s removal. The finding suggests that regret aversion bias may influence investment decisions, but its impact can differ depending on various study-specific factors.
4.2.1.3 Overconfidence bias and investment decisions
The meta-analysis includes 23 studies exploring the correlation between overconfidence bias and investment decision-making. The overall estimate of the relationship is 0.346 with an S.E. of 0.0949. The Z-statistic is 3.65, and the p-value is < 0.001, indicating a statistically significant relationship. Similar to the previous hypotheses, there is significant heterogeneity among the studies, with an I2 value of 0.9847 and a Q-test value of 920.599 (p < 0.001), indicating substantial heterogeneity. Removing the outlier from the dataset resulted in a reduction in the estimate (0.283) with a standard error of 0.0728. The Z-statistic of 3.98 and p-value <0.001 also indicate a highly significant relationship even after the outlier’s removal. This means that while overconfidence bias is significantly related to investment decisions, the effect sizes vary widely across the studies. This variability may be influenced by different methodologies, sample characteristics, or other factors.
4.2.2 Heterogeneity
Table 3 shows significant variability and diversity among the studies that are included in this analysis. This high I2 value indicates substantial variability among the studies, which implies that emotional biases affect the true effect size differently in different studies. The Q-tests for heterogeneity are all highly significant (p < 0.001), also confirming the presence of significant variability among the study results. These findings suggest that the true effects of these biases may differ considerably across different studies, maybe because of their population, sample, or investment choices.
All three hypotheses were supported, as there were statistically significant positive relationships between emotional biases (loss aversion, regret aversion, and overconfidence) and investment decisions. However, the presence of high heterogeneity among the studies indicates that additional factors may influence the effect sizes observed.
4.2.3 Forest plot
4.2.3.1 Loss aversion bias and investment decisions
The forest plot in Figure 1 shows a substantial positive relation between loss aversion bias and investment decisions with 1 possible outlier (RLB-1) (See Figure 1). This means that loss-averse investors are more likely to make decisions that avoid losses, even if those decisions are not in their best interests overall. The confidence interval for the average effect size is 0.10–0.88, which means that we can be 95% confident that the actual effect size falls within this range.
4.2.3.2 Regret aversion bias and investment decisions
The forest plot in Figure 2 exhibits a significant correlation between regret aversion bias and investment decision-making with 1 possible outlier (OR-5) (See Figure 2). The overall estimate of the effect size is 0.40, which means that regret aversion bias is also significantly correlated with investment decisions. The confidence interval for the overall estimate is 0.14–0.66, which means that we can be 95% confident that the actual effect size falls within this range.
4.2.3.3 Overconfidence bias and investment decisions
The forest plot in Figure 3 shows a significant correlation between overconfidence bias and investment decisions, with 1 possible outlier (OR-5) (See Figure 3). The overall estimate of the effect size is 0.35, which means that overconfidence bias is positively correlated with investment decisions. The confidence interval for the overall estimate is 0.16–0.53, which means that we can be 95% confident that the actual effect size falls within this range.
4.2.4 Publication bias
Table 4 shows the result of the publication bias tests suggests that there is no significant evidence of publication bias for any of the bias types studied (See Table 4). The p-values for all the variables (loss aversion bias, loss aversion bias w/o outlier, regret aversion bias, regret aversion bias w/o outlier, overconfidence bias, and overconfidence bias w/o outlier) are less than 0.001, indicating strong evidence against publication bias. Furthermore, the “Fail-Safe N” values are large enough, which means that a substantial number of non-significant studies would be required to change the overall conclusion. The “Begg and Mazumdar Rank Correlation” values are all close to 0 and not statistically significant, indicating no significant correlation between study effect sizes and their variances, which is another indicator of a lack of publication bias.
5. Discussions and conclusions
5.1 General interpretation
The results of this meta-analytical study show a significant positive correlation between emotional biases (loss aversion bias, regret aversion bias, and overconfidence bias) and investment decisions. The meta-analysis approach allowed us a more thorough understanding of how these emotional biases impact investment decisions by combining data from multiple independent studies. These findings align with previous studies, providing further extra support for the understanding that human emotions play a significant role in shaping investment decisions.
This meta-analytical study revealed the following key findings:
Loss aversion bias: a positive relationship with an investment decision with a correlation value of 0.492 is exhibited in Figure 4 (See Figure 4). Investors who show a loss-aversion bias tend to be more risk-averse and prioritise avoiding losses over capital growth. This bias leads them to hold onto declining assets in the hope of recovering losses and sell winning assets prematurely to secure gains. However, the effect sizes vary widely across studies, indicating that the impact of loss aversion on investment decisions can differ depending on various factors and various populations.
Regret aversion bias: Figure 4 further exhibits a positive relationship with investment decisions with a correlation of 0.401 (See Figure 4). Investors with a regret aversion bias tend to be overly conservative, avoiding risky decisions due to fear of regretting potential losses. This bias can lead to missed opportunities in undervalued markets and holding onto losing positions. Again, the effect sizes vary among studies, suggesting that the influence of regret aversion on investment decisions can differ in different contexts.
Overconfidence bias: The meta-analysis also demonstrates a positive relationship between overconfidence bias and investment decision-making. Overconfident investors overestimate their abilities, leading to ill-advised purchases and a reluctance to sell their holdings. This bias can also lead to excessive trading and under-diversification of portfolios, resulting in subpar long-term returns. Similar to the other biases, the effect sizes vary significantly across the studies, also indicating that the impact of overconfidence on investment decisions can be dependent on context.
5.2 Limitations of the evidence included in the review
Despite the overall consistency in the relationships between emotional biases and investment decisions, the evidence included in this review is not free from limitations. One notable limitation is the heterogeneity observed among the studies. The considerable variability in effect sizes may be due to differences in methodologies, sample characteristics, population, and other cultural contexts. This heterogeneity also suggests that the strength of these relationships may vary across different populations and circumstances.
Another limitation lies in the use of correlational data from the included studies. While correlations provide valuable insights into the relationships between variables, they do not establish causation. The observed associations between emotional biases and investment decisions do not necessarily imply a direct causal relationship. Other factors, such as individual risk tolerance, financial literacy, and market conditions, may also influence investment decisions, and their interactions with emotional biases should be further explored.
Additionally, the limited number of studies exploring regret-aversion bias and its impact on investment decisions may affect the robustness of the results. A larger body of research on this specific emotional bias could strengthen the generalisability of the findings.
5.3 Limitations of the review processes
The review processes used in this meta-analysis may also introduce certain limitations. One potential limitation is the possibility of publication bias. Efforts were made to minimise this bias by conducting an extensive literature search and including studies from various sources, including unpublished or non-significant studies. Publication bias could impact the validity of the results if it disproportionately favours studies with significant effects, potentially overestimating the overall effect size.
Another potential limitation is the use of correlational data from the included studies. Meta-analytic studies generally rely on available data from published sources, which may restrict the availability of other types of data, such as experimental or longitudinal designs. The reliance on correlations limits the ability to draw causal conclusions, as mentioned earlier.
Furthermore, the exclusion of qualitative, viewpoint, and systematic literature reviews may lead to a potential bias in the selection of studies. While these types of studies might not provide correlation data, they can offer valuable insights and interpretations of the relationship between these emotional biases and investment decisions.
5.4 Implications of the results
The findings of this meta-analysis have significant implications for individual investors, as 80% of our life decisions are based on emotion and emotional intelligence (Cotruş et al., 2012). Understanding the influence of emotional biases on investment decisions might help investors make more informed and rational decisions while managing their portfolios. Investors should be more conscious of their own emotional biases and take the necessary actions to reduce their influence on decision-making. Diversification, setting clear financial goals, and getting advice from a financial advisor can also help to reduce the influence of these biases on investment decisions.
Figures
Study characteristics
Author and year | Country | OCB | LAB | RAB | Sample | DV | IV | Code |
---|---|---|---|---|---|---|---|---|
Bhatia et al. (2022) | India | 0.375 | 0.259 | 172 | OCB, LAB | ID | OL-1 | |
Bihari et al. (2023) | India | 0.267 | 0.237 | 0.406 | 337 | OCB, LAB, RAB | ID | OLR-1 |
Lambert et al. (2012) | France | −0.02 | 84 | OCB | ID | OC-1 | ||
Wangzhou et al. (2021) | Pakistan | 0.042 | 200 | RAB | ID | RAB-1 | ||
Rahman and Gan (2020) | Malaysia | 0.49 | 502 | OCB | ID | OC-2 | ||
Gupta and Shrivastava (2022) | India | 0.584 | 119 | LAB | ID | LAB-1 | ||
Kasoga (2021) | Tanzania | 0.645 | 216 | OCB | ID | OC-3 | ||
Shah et al. (2018) | Pakistan | −0.294 | 183 | OCB | ID | OC-4 | ||
Baker et al. (2019) | India | 0.615 | 253 | OCB | ID | OC-5 | ||
Ahmad and Shah (2022) | Pakistan | −0.202 | 183 | OCB | ID | OC-6 | ||
Jan et al. (2022) | China | 0.345* | 1,000 | OCB | ID | OC-7 | ||
Thevaruban (2022) | Sri Lanka | −0.034 | 0.073 | 165 | OCB, RAB | ID | OR-1 | |
Octavia et al. (2022) | Indonesia | 0.546 | 79 | RAB | ID | RAB-2 | ||
Iram et al. (2021) | Pakistan | 0.986 | 0.453 | 579 | RAB, LAB | ID | RLB-1 | |
Gyawali and Neupane (2021) | Nepal | 0.392* | 0.327* | −0.003* | 347 | OCB, LAB, RAB | ID | OLR-2 |
Dhungana et al. (2022) | Nepal | 0.446 | 0.325 | 179 | OCB, RAB | ID | OR-2 | |
Sapkota (2023) | Nepal | 0.531 | 0.392 | 0.451 | 385 | OCB, LAB, RAB | ID | OLR-3 |
Rehan and Umer (2017) | Pakistan | 0.09 | 0.38 | 385 | OCB, RAB | ID | OR-3 | |
Hunguru et al. (2020) | Zimbabwe | 0.539 | 0.422 | 291 | RAB, LAB | ID | RLB-2 | |
Kumar et al. (2018) | India | 0.631* | 116 | LAB | ID | LAB-2 | ||
Laungratanamas and Nuangjamnong (2022) | Thailand | 0.43* | 411 | OCB | ID | OC-8 | ||
Chadha (2024) | India | 0.117 | 0.18 | 1,012 | OCB, RAB | ID | OR-4 | |
Elhussein and Abdelgadir (2020) | Sudan | 0.508* | 0.201* | 0.128* | 203 | OCB, LAB, RAB | ID | OLR-4 |
Armansyah (2021) | Indonesia | 0.435 | 0.075 | 250 | LAB | ID | OL-2 | |
Aydin (2023) | Turkey | −0.169 | 352 | LAB | ID | LAB-3 | ||
Michael and Oshoma (2022) | Nigeria | 0.589 | 308 | OCB | ID | OC-9 | ||
Benayad and Aasri (2023) | Moroccan | −0.06 | 133 | OCB | ID | OC-10 | ||
Athur (2014) | Kenya | 0.003 | −0.022 | 31 | RAB, LAB | ID | RLB-3 | |
Nkukpornu et al. (2020) | Ghana | 0.946 | 0.964 | 120 | OCB, RAB | ID | OR-5 | |
Hassan Metwally (2023) | Egypt | −0.427 | 384 | OCB | ID | OC-11 | ||
Michael (2023) | Nigeria | 0.343* | 340 | OCB | ID | OC-12 |
Note(s): *Conversion of Beta into Correlation coefficients. (Refer− 3.3)
Source(s): Authors’ work
Correlation coefficient table of emotional biases
Independent variable | No. of study | Estimate | S. E | Z | p | CI lower Bound | CI Upper Bound |
---|---|---|---|---|---|---|---|
Loss aversion bias | 12 | 0.492 | 0.198 | 2.48 | 0.013 | 0.103 | 0.881 |
Loss aversion bias after removal of outlier | 11 | 0.312 | 0.0848 | 3.68 | <0.001 | 0.146 | 0.478 |
Regret aversion bias | 14 | 0.401 | 0.134 | 3 | 0.003 | 0.139 | 0.664 |
Regret aversion bias after removal of outlier | 13 | 0.285 | 0.0576 | 4.96 | <0.001 | 0.173 | 0.398 |
Overconfidence bias | 23 | 0.346 | 0.0949 | 3.65 | <0.001 | 0.161 | 0.532 |
Overconfidence bias after removal of outlier | 22 | 0.283 | 0.0728 | 3.89 | <0.001 | 0.14 | 0.426 |
Source(s): Authors’ work
Heterogeneity statistics table (emotional biases)
Independent variable | I2 | Q | p |
---|---|---|---|
Loss aversion bias | 0.9917 | 2424.159 | <0.001 |
Loss aversion bias after removal of outlier | 0.943 | 164.815 | <0.001 |
Regret aversion bias | 0.9861 | 463.639 | <0.001 |
Regret aversion bias after removal of outlier | 0.9193 | 133.141 | <0.001 |
Overconfidence bias | 0.9847 | 920.599 | <0.001 |
Overconfidence bias after removal of outlier | 0.9735 | 663.388 | <0.001 |
Source(s): Authors’ work
Publication bias
Independent variable | Fail-Safe N | Begg and Mazumdar Rank correlation | ||
---|---|---|---|---|
Value | p | Value | p | |
Loss aversion bias | 4,294 | <0.001 | 0.212 | 0.381 |
Loss aversion bias after removal of outlier | 858 | <0.001 | 0.091 | 0.761 |
Regret aversion bias | 2,654 | <0.001 | 0.022 | 0.913 |
Regret aversion bias after removal of outlier | 1,469 | <0.001 | −0.09 | 0.669 |
Overconfidence bias | 6,453 | <0.001 | 0.012 | 0.937 |
Overconfidence bias after removal of outlier | 4,696 | <0.001 | −0.091 | 0.553 |
Source(s): Authors’ work
Conflict of interest statement: The author(s) hereby declare that there is no conflict of interest.
References
Adil, M., Singh, Y. and Ansari, M.S. (2022), “How financial literacy moderate the association between behaviour biases and investment decision?”, Asian Journal of Accounting Research, Vol. 7 No. 1, pp. 17-30, doi: 10.1108/AJAR-09-2020-0086.
Ahmad, M. and Shah, S.Z.A. (2022), “Overconfidence heuristic-driven bias in investment decision-making and performance: mediating effects of risk perception and moderating effects of financial literacy”, Journal of Economic and Administrative Sciences, Vol. 38 No. 1, pp. 60-90, doi: 10.1108/jeas-07-2020-0116.
Armansyah, R.F. (2021), “Over confidence, mental accounting, and loss aversion in investment decision”, Journal of Auditing, Finance, and Forensic Accounting, Vol. 9 No. 1, pp. 44-53, doi: 10.21107/jaffa.v9i1.10523.
Athur, A.D. (2014), Effect of Behavioural Biases on Investment Decisions of Individual Investors in Kenya, University of Nairobi, available at: http://erepository.uonbi.ac.ke/handle/11295/74832 (accessed 18 January 2024).
Aydin, S. (2023), “Analysis of the effects of loss aversion, optimism and herding biases and fear of missing out on investment decision”, available at: https://www.researchgate.net/publication/375187605
Baker, H.K., Kumar, S., Goyal, N. and Gaur, V. (2019), “How financial literacy and demographic variables relate to behavioral biases”, Managerial Finance, Vol. 45 No. 1, pp. 124-146, doi: 10.1108/MF-01-2018-0003.
Baker, H.K., Kumar, S. and Goyal, N. (2021), “Personality traits and investor sentiment”, Review of Behavioral Finance, Vol. 13 No. 4, pp. 354-369, doi: 10.1108/RBF-08-2017-0077.
Benayad, K. and Aasri, M.R. (2023), “Behavioral biases and investment decisions of SMEs managers: empirical analysis within the Moroccan context”, International Journal of Financial Studies, Vol. 11 No. 4, p. 120, doi: 10.3390/ijfs11040120.
Bhatia, A., Chandani, A., Divekar, R., Mehta, M. and Vijay, N. (2022), “Digital innovation in wealth management landscape: the moderating role of robo advisors in behavioural biases and investment decision-making”, International Journal of Innovation Science, Vol. 14 Nos 3-4, pp. 693-712, doi: 10.1108/IJIS-10-2020-0245.
Bihari, A., Dash, M., Muduli, K., Kumar, A., Mulat-Weldemeskel, E. and Luthra, S. (2023), “Does cognitive biased knowledge influence investor decisions? An empirical investigation using machine learning and artificial neural network”, VINE Journal of Information and Knowledge Management Systems, Vol. ahead-of-print No. ahead-of-print, doi: 10.1108/VJIKMS-08-2022-0253.
Borenstein, M., Hedges, L.V., Higgins, J.P.T. and Rothstein, H.R. (2009), Introduction to Meta-Analysis, Wiley, doi: 10.1002/9780470743386.
Chadha, M. (2024), “Cognitive biases impact on investor's investment decision of retail mutual fund investor”, European Economic Letters (EEL), Vol. 14 No. 1, pp. 1989-2002, doi: 10.52783/eel.v14i1.1131.
Cochran, W.G. (1954), “The combination of estimates from different experiments”, Vol. 10 No. 1, p. 101, doi: 10.2307/3001666, available at: https://www.jstor.org/stable/3001666?seq=1&cid=pdf
Cotruş, A., Stanciu, C. and Bulborea, A.A. (2012), “EQ vs. IQ which is most important in the success or failure of a student?”, Procedia - Social and Behavioral Sciences, Vol. 46, pp. 5211-5213, doi: 10.1016/j.sbspro.2012.06.411.
Dhungana, B.R., Bhandari, S., Ojha, D. and Sharma, L.K. (2022), “Effect of cognitive biases on investment decision making: a case of Pokhara valley, Nepal”, Quest Journal of Management and Social Sciences, Vol. 4 No. 1, pp. 71-84, doi: 10.3126/qjmss.v4i1.45868.
Elhussein, N.H.A. and Abdelgadir, J.N.A. (2020), “Behavioral bias in individual investment decisions: is it a common phenomenon in stock markets?”, International Journal of Financial Research, Vol. 11 No. 6, p. 25, doi: 10.5430/ijfr.v11n6p25.
Gupta, S. and Shrivastava, M. (2022), “Herding and loss aversion in stock markets: mediating role of fear of missing out (FOMO) in retail investors”, International Journal of Emerging Markets, Vol. 17 No. 7, pp. 1720-1737, doi: 10.1108/IJOEM-08-2020-0933.
Gyawali, I. and Neupane, G. (2021), “Individual investors psychology and investment decision in NEPSE”, The Lumbini Journal of Business and Economics, Vol. 9 No. 1-2, pp. 43-53, doi: 10.3126/ljbe.v9i1-2.45986.
Hansen, C., Steinmetz, H. and Block, J. (2022), “How to conduct a meta-analysis in eight steps: a practical guide”, Management Review Quarterly, Vol. 72 No. 1, pp. 1-19, doi: 10.1007/s11301-021-00247-4.
Hassan Metwally, A. (2023), “The effect of overconfidence bias on investors decisions in the Egyptian stock market: the role of information acquisition 1 the effect of overconfidence bias on investors decisions in the Egyptian stock market [48]”, Journal of Alexandria University for Administrative Sciences©, Vol. 60 No. 2, pp. 47-85, doi: 10.21608/acj.2023.294144.
Hunguru, P., Sibanda, V. and Tadu, R. (2020), “Determinants of investment decisions: a study of individual investors on the Zimbabwe stock exchange”, Applied Economics and Finance, Vol. 7 No. 5, p. 38, doi: 10.11114/aef.v7i5.4927.
Iram, T., Bilal, A.R. and Latif, S. (2021), “Is awareness that powerful? Women’s financial literacy support to prospects behaviour in prudent decision-making”, Global Business Review, doi: 10.1177/0972150921996185.
Jan, N., Jain, V., Li, Z., Sattar, J. and Tongkachok, K. (2022), “Post-COVID-19 investor psychology and individual investment decision: a moderating role of information availability”, Frontiers in Psychology, Vol. 13, 846088, doi: 10.3389/fpsyg.2022.846088.
Kahneman, D. and Tversky, A. (1979), “Prospect theory: an analysis of decision under risk”, Econometrica, Vol. 47 No. 2, p. 263, doi: 10.2307/1914185.
Kasoga, P.S. (2021), “Heuristic biases and investment decisions: multiple mediation mechanisms of risk tolerance and financial literacy—a survey at the Tanzania stock market”, Journal of Money and Business, Vol. 1 No. 2, pp. 102-116, doi: 10.1108/jmb-10-2021-0037.
Kumar, A.A., Babu, D.M. and Scholar, P. (2018), Effect of Loss Aversion Bias on Investment Decision: A Study, JETIR, Vol. 5, available at: www.jetir.org
Lambert, J., Bessière, V. and N'Goala, G. (2012), “Does expertise influence the impact of overconfidence on judgment, valuation and investment decision?”, Journal of Economic Psychology, Vol. 33 No. 6, pp. 1115-1128, doi: 10.1016/j.joep.2012.07.007.
Laungratanamas, K. and Nuangjamnong, C. (2022), “Biases and fear of missing out impact investment decisions in Thailand during COVID-19 pandemic international journal of advanced multidisciplinary research and studies behavioral biases and fear of missing out impact investment decisions in Thailand during COVID-19 pandemic”, available at: www.multiresearchjournal.com
Michael, C.I. (2023), “Behavioural biases and investment decisions in the Nigerian stock exchange behavioural biases and investment decisions in Nigerian stock exchange”, Vol. 5 No. 4, pp. 2023-2133, available at: http://journals.uonbi.ac.ke/index.php/adfjhttp://journals.uonbi.ac.ke/index.php/adfj
Michael, C.I. and Oshoma, O.B. (2022), “Examining the influence of anchoring bias and overconfidence on investment decision-making in the Nigerian exchange group: a case study of investors in Edo state”, Nigeria, Vol. 5 No. 4, pp. 2023-2133, available at: http://journals.uonbi.ac.ke/index.php/adfjhttp://journals.uonbi.ac.ke/index.php/adfj
Nkukpornu, E., Gyimah, P. and Sakyiwaa, L. (2020), “Behavioural finance and investment decisions: does behavioral bias matter?”, International Business Research, Vol. 13 No. 11, p. 65, doi: 10.5539/ibr.v13n11p65.
Octavia, A.D., Yuwono, C.C. and Monangin, F.G. (2022), “The effect of financial literature, financial behavior, & regret aversion bias on millennial investment decisions”, Indikator: Jurnal Ilmiah Manajemen Dan Bisnis, Vol. 6 No. 3, p. 82, doi: 10.22441/indikator.v6i3.14219.
Peterson, R.A. and Brown, S.P. (2005), “On the use of beta coefficients in meta-analysis”, Journal of Applied Psychology, Vol. 90 No. 1, pp. 175-181, doi: 10.1037/0021-9010.90.1.175.
Pompian, M. (2012), Behavioral Finance and Wealth Management, in Pompian, M. (Ed.), Wiley, doi: 10.1002/9781119202400.
Raheja, S. and Dhiman, B. (2020), “How do emotional intelligence and behavioral biases of investors determine their investment decisions?”, Rajagiri Management Journal, Vol. 14 No. 1, pp. 35-47, doi: 10.1108/RAMJ-12-2019-0027.
Rahman, M. and Gan, S.S. (2020), “Generation Y investment decision: an analysis using behavioural factors”, Managerial Finance, Vol. 46 No. 8, pp. 1023-1041, doi: 10.1108/MF-10-2018-0534.
Rehan, R. and Umer, I. (2017), “Market forces behavioural biases and investor decisions”, Market Forces Research Journal, Vol. 12 No. 2, available at: https://kiet.edu.pk/marketforces/index.php/marketforces/article/view/341 (accessed 18 January 2024).
Ritika and Kishor, N. (2022), “Development and validation of behavioral biases scale: a SEM approach”, Review of Behavioral Finance, Vol. 14 No. 2, pp. 237-259, doi: 10.1108/RBF-05-2020-0087.
Sapkota, M.P. (2023), “Emotional biases and equity investment decision of individual investors”, Journal of Business and Management Review, Vol. 4 No. 1, pp. 036-049, doi: 10.47153/jbmr41.5682023.
Shah, S.Z.A., Ahmad, M. and Mahmood, F. (2018), “Heuristic biases in investment decision-making and perceived market efficiency: a survey at the Pakistan stock exchange”, Qualitative Research in Financial Markets, Vol. 10 No. 1, pp. 85-110, doi: 10.1108/QRFM-04-2017-0033.
Thevaruban, J.S. (2022), “The impact of behavioural factors on individual investors' investment decision making in colombo stock exchange: special reference to Northern province-Sri Lanka”, in International Journal of Research and Innovation in Social Science, available at: www.rsisinternational.org
Tversky, A. and Kahneman, D. (1974), “Judgment under uncertainty: heuristics and biases”, In New Series, Vol. 185 No. 4157.
Viechtbauer, W. (2010), “Conducting meta-analyses in R with the metafor package”, Journal of Statistical Software, Vol. 36 No. 3, doi: 10.18637/jss.v036.i03.
Wangzhou, K., Khan, M., Hussain, S., Ishfaq, M. and Farooqi, R. (2021), “Effect of regret aversion and information cascade on investment decisions in the real estate sector: the mediating role of risk perception and the moderating effect of financial literacy”, Frontiers in Psychology, Vol. 12, 736753, doi: 10.3389/fpsyg.2021.736753.
Further reading
Geyskens, I., Krishnan, R., Steenkamp, J.-B.E.M. and Cunha, P.V. (2009), “A review and evaluation of meta-analysis practices in management research”, Journal of Management, Vol. 35 No. 2, pp. 393-419, doi: 10.1177/0149206308328501.
Huedo-Medina, T.B., Sánchez-Meca, J., Marín-Martínez, F. and Botella, J. (2006), “Assessing heterogeneity in meta-analysis: Q statistic or I 2 Index?”, Psychological Methods, Vol. 11 No. 2, pp. 193-206, doi: 10.1037/1082-989X.11.2.193.
Ullah, S., Elahi, M.A., Ullah, A., Pinglu, C. and Subhani, B.H. (2020), “Behavioral biases in investment decision making and moderating role of investor's type”, Intellectual Economics, Vol. 14 No. 2, pp. 87-105, doi: 10.13165/IE-20-14-2-06.
Zhang, X. (2023), “The role of behavioral bias in investment outcomes”, In Business, Economics and Management FTMM, Vol. 2023.
Corresponding author
About the authors
Shailendra Kumar is an Assistant Professor in the Department of Management, Sikkim University, Gangtok, India. Business ethics, corporate social responsibility, and artificial intelligence are among his research and specialisation areas. He has written extensively on the topic of Artificial Intelligence and social robots. Dr Kumar is a notable Tech. Philosopher and Researcher. When it comes to AI, he has been an early trailblazer in terms of thought. In AI coding and programming, he has been an early advocate of drawing on the normative notions of classical philosophers. His innovative arguments on the identity of AI humanoids as artificial persons and legal entities are a significant addition to the literature on AI ethics and philosophy.
Akash Chaurasia is a PhD scholar at the Department of Management, Sikkim University.