BEHAVIORAL FACTORS INFLUENCING INVESTMENT DECISIONS: A SYSTEMATIC REVIEW USING PRISMA

The purpose of this systematic review is to analyze the previous study made in the past related to an association between behavioral biases and investment decisions. This systematic review article identified 21 distinct biases affecting the investment decisions made by investors. Data were collected from 3 databases (Scopus, Taylor and Francis, and Web of Science) using PRISMA guidelines. The inclusion criteria for this systematic review were (i) publication date 2012 to 2022 (ii) published in the English language (iii) open access articles (iv) all types of documents except review articles etc. The findings of the study support that there is a significant relationship between 21 distinct behavioral biases and investment decisions. The study found there is a lack of female participants to male participants. There is a sampling bias found in many studies i.e., the studies employed non-probability techniques to collect samples. This study extends the current knowledge about the previous study being researched related to an association between behavioral biases and investment decisions. This systematic review provides a clear picture of the gaps in the previous literature. This study is helpful for researchers and academicians in understanding the biases investors undergo while making investment decisions and addressing the gaps identified related to different types of behavioral biases and areas of study that need to be explored more in near future.


INTRODUCTION
Various research studies identified that investors do not behave always rationally while making investment decisions.With this idea, a new concept i.e. behavioral finance emerged in the area of finance and economics in the 1980s.Irrationality in human behavior can be found in the form of distinct behavioral biases and compares behavioral finance theories with traditional and modern finance theories.Nair and Antony (2015) viewed behavioral finance as not only a replacement for classical finance theories but also as a means to understand the irrational behavior of investors and the reasons for sudden growth and decline in the market.Behavioral finance proposes that the investment decision-making process is influenced by various behavioral biases that boost investors to deviate from rationality and take irrational investment decisions (Niehaus & Shrider, 2014).The present study is an extensive review of behavioral biases in individual investment decision-making (Taffler, Spence, & Eshraghi, 2017).After studying various pieces of literature, it was found that there is a need to conduct one study that carries a systematic review of behavioral biases (Kumar & Goyal, 2015).This made us curious to identify various biases studied in past.This systematic review article made an effort in identifying distinct behavioral biases and their influence on the investment decisions of investors.
The remaining study is structured as section 2 explains the method used in this systematic review, section 3 discussed the results of the study, section 4 deals with discussion, section 5 is about the limitations of this systematic review, and section 6 deals with the conclusion of the study.

MATERIALS AND METHODS Eligibility Criteria
All studies assessing the relationship between behavioral actors and investment decisions were eligible for this systematic review.The inclusion criteria were (i) Publication between 2012 to May 2022 (ii) Written in English (iii) Conducted an assessment of behavioral biases and investment decisions objectively (iv) Only open-access studies included (v) All types of documents, research papers, articles, etc. (vi) Subject included: Accounting, Business, Business Finance, Business Industry, Economics, Econometrics & Finance, and Finance & Management (vii) Database: Scopus, Taylor and Francis, and Web of Science.

Basis Inclusion Exclusion Database
Scopus, Web of Science, and Taylor and Francis All the databases except those mentioned in the inclusion

Study topic
Related to the association between biases and investment decisions.

Study Language
Only English Other than the English language Time frame of the study

Published from 2012 to May 2022
Published before 2012.

Type of document
Research papers, articles, conference papers, working papers, etc.

Subject area
Accounting, Business, Business Finance, Business Industry, Economics, Econometrics & Finance, and Finance and Management Other than mentioned in the inclusion Previous studies were excluded from the systematic review if they were (i) assess the relationship between behavioral biases and investment decisions with some other factors (ii) review articles (iii) not published in peer-reviewed journals (iv) language other than English.

Information Source and Search
A literature search for the systematic review article was conducted in May 2022 including the databases viz., Scopus, Taylor and Francis, and Web of Science.Numerous searchers were conducted in the above-stated electronic databases and after performing various trials and errors the final search term for the study: "Behavioral Factors" OR "Behavioral Biases" OR "Cognitive Biases" OR "Psychological Factors" AND "Invest* Decision"

Study Selection and Data Collection Processes
After performing the initial literature searches through various academic databases based on title, abstract, and keywords, the relevant studies were further assessed and screened considering the eligibility criteria for inclusion of the study in this systematic review.Detailed information about the selection of studies in this systematic review was presented in the PRISMA flow diagram (Figure 1).

Study Selection
A total number of 700 studies were identified in the process of initial search (Scopus,n= 350;Taylor and Francis,n= 168,and Web of Science,n= 182).After investigating the title, abstract, and keywords of all studies being searched through various databases (n=663) studies were excluded from the systematic review due to unsuitability for this study.Further, (n=4) studies were removed due to duplicates.As a result (n=33) studies were selected for the full-text eligibility phase of the study.Out of the studies selected for the eligibility of full text (n=4), studies were excluded due to not assessing the relationship between behavioral factors and investment decisions.Finally, (n=29) studies were selected for systematic review.

Study Characteristics
Information related to the general and specific properties of all included Studies (n=29) can be found in Tables 1, 2, and 3  The results of the study show that amongst male investors, the influence of herding and risk-aversion on investors' investment decisions was negative and statistically significant, whereas the influence of overconfidence on investment decisions was positive and significant.However, the effect of disposition was found statistically insignificant.Researchers also found that amongst female investors the effect of herding and risk-aversion on investment decisions was the same as that of males.However, the influence of the disposition effect and overconfidence was statistically insignificant on investment decisions.The moderation effect of financial literacy between overconfidence and investment decision amongst male investors was significantly significant.However, the moderation of financial literacy with the remaining three biases was found insignificant.The results of moderation of financial literacy with overconfidence, herding, disposition, and risk-aversion were found statistically significant amongst female investors.The findings show that female and mature investors suffer weak disposition bias as compared to male and young investors.The result also shows that the disposition effect is more pronounced in a bull market than in a bear market.

PLS-SEM
Loss-aversion has a positive and significant effect on investment decisions.Negative and significant influence was found in the case of herding as an intervening variable between loss-aversion and investment decisions.No significant association was found between overconfidence and investment decisions.Researchers also concluded that investment in the real assets sector required more complicated decisions than financial assets.Based on the types of investors the result reflects that active investor shows overconfidence bias whereas passive investors show herding bias while making investment decisions.The result found that overconfidence and herding exist in KSE whereas the disposition effect does not exist in KSE.Researchers also identify that all three biases have a positive and significant relationship with investment decision-making.All the biases significantly impacted the investment decisions made by investors.Investors' type shows no moderating contribution between disposition effect and investment decision, negative moderating contribution between herding and investment decisions, and positive moderating contribution between overconfidence and investment decisions.The study found a positive impact of representative and availability biases on investors' investment decisions.No significant impact of psychological bias was found on investment decisions.Researchers could not find any moderating role of locus of control between any biases and investment decisions.

SEM
Researchers found that there is a positive and significant effect of overconfidence on stock investment decisions.It is also found that there is no significant relationship between overconfidence and investment decisions by taking the year of investment as a moderating variable.

Descriptive statistics, Correlation
The study found that overconfidence is the most pronounced bias found among retail investors in the Bhubaneswar region of Odisha.The study also found that investors use media reporting while making investment decisions and they are also influenced by the disposition effect.

Correlation, Regression Analysis
The result shows that herding and overconfidence significantly influence the investment decisions of investors whereas the disposition effect and anchoring do not have a significant impact on investors' investment decisionmaking.

Secondary data
Quantile Regression Model, descriptive statistic, GARCH model The researchers did not identify the presence of herding in extreme market conditions amid the pandemic when full market return data were used by investors.Herding was detected during the pandemic in banking and financial services during the bull market conditions.The study found that while making investment decisions, investors are prone to these behavioral biases.Only herding has a significant relationship with one of the components of investment decision-making which is demand identification, while information search shows a strong positive association with mental accounting and optimism.Additionally, researchers also found that only the disposition effect has a significant relationship with evaluating alternative components of the decision-making process.The result of both quantitative studies as well as qualitative studies indicates that all biases except over-optimism exist in investment decisions made by investors.As per order herding is the most influential, then hindsight, then overconfidence, then representativeness, then mental accounting, and the other two biases Viz., anchoring, and loss-aversion slightly influence the trading activity of investors.The same pattern follows in the case of recommendation intentions also.The study also reveals that the influence of biases is more in their trading activity decisions compared to recommendation intentions to others.A post hoc study indicates that mental accounting and lossaversion have a substantial influence on recommendation intention.The qualitative study also reveals that most of the biases observed at the beginning of the pandemic continue with its advancement in millennials' investment decisions.The results are aligned with previous results but are more robust than the disposition effect present in the financial market.Trading activity and financial literacy slightly mitigate the disposition effect.

Multiple Regression
Investors in the mutual fund markets exhibit the disposition effect.In the Bull market stock fund, investors show the disposition effect whereas, in the Bear market, neither stock fund investors nor balanced fund investors exhibit the disposition effect.Researchers also found that in the Neutral market overall all types of investors exhibit the disposition effect.

Types of Biases
Regarding the types of biases, this systematic review identifies various types of biases the previous studies deal with or study about.There were approximately 21 types of biases that were previously studied by various researchers and academicians.Most of the previous studies dealt with overconfidence and herding bias.

Participants
This systematic review study included a total number of 6,802 participants from the primary source of data collection and the majority of the population included in the previous studies were male than female samples.All the studies included adult samples apart from a study that includes both adult as well as student samples for their study (Rzeszutek, Szyszka, & Czerwonka, 2015).One of the studies uses the composition of consultants, sales and purchase brokers, secretary general, director, and research specialist (Akgul & Cetin, 2021).One study also includes small, medium, and professional investors of different brokerage firms (Hafez, 2021).One study also comprises bankers with individual investors as a sample of their study (Qasim, Hussain, Mehboob, & Arshad, 2019).A study recruited only millennial male participants in their study (Talwar, Talwar, Tarjanne, & Dhir, 2021).The participants of one of the studies were Russian speakers around the world (Zhdanov & Simonov, 2021).

Methodological Features
Regarding methodological features being adapted from previous studies, this systematic review article identified that most of the previous studies were descriptive research and used quantitative methods.Some of the studies also used crosssectional data for the attainment of the objectives of the research.Structured questionnaires were used by researchers for collecting the data and measuring hypotheses objectively.Most of the studies used correlation and regression for the analysis of results.Previous studies also used Structure Equation Modeling (SEM).Secondary data-based studies used various models for the analysis of results.

Risk of Bias
Concerning the procedure of assessing the risk of sampling bias, approximately all the previous studies were found biased in the case of sampling, as they follow non-probability sampling techniques for collecting their samples.Most of the primary data-based studies use snowball, convenient, purposive sampling techniques for collecting data.One of the studies used experimental methods in form of investment games for collecting data (Zhdanov & Simonov, 2021).One study used only male samples and female respondents were completely ignored (Talwar, Talwar, Tarjanne, & Dhir, 2021).

DISCUSSIONS
The present systematic reviews article various peer-reviewed published research papers investigating the association between behavioral biases and investment decisions.All the studies included in this systematic review article reflects the presence of behavioral biases while making investment decision by investors and found more or less every investor are prone to various biases in the process of investment decision-making.This systematic review also extracted data from reviewed studies including (i) data collected from countries (ii) context (iii) types of biases deals (iv) characteristics of participants (v) information regarding sample (vi) risk of bias, and (vii) methodological features.
In respect of geographical dispersion, (n=21) studies were conducted in parts of Asia, and the rest of the studies were in different parts of the world.In terms of the context of the market, the previous studies deals (n=19) studies deal with the stock market/equity market/capital market and the rest of the studies deal with cryptocurrency, mutual funds, real assets, banking, and financial services, and different investment avenues available for investment.With regards to the types of biases in the previous studies, the majority of the studies (n=26) investigate the relationship between overconfidence, herding, and disposition effect with investment decisions as combined or individually or by combining with other biases.Other types of biases dealt with in the previous studies are gambling effect, risk-aversion, Desirability bias, endowment effect, anchoring, status quo, Bandwagon effect, home bias, Availability, loss-aversion, loss-aversion, psychological bias, excessive optimism (pessimism), certainty effect, the sunk cost fallacy, and mental accounting, hindsight and Familiarity Bias.More specifically, the previous research comprised fewer female participants than males, individual investors of their respective markets rather than experts in the field.One of studies used only male participants and ignore the female CONCLUSIONS As a concluding remark, the findings obtained from this systematic review make evidence that the investment decisions of investors are affected by various types of biases.There are a lot of works mainly on three types of biases viz., overconfidence, herding, and the disposition effect.A lot of work has been done in the context of behavioral biases affecting the investment decisions of investors in the capital market/ stock market/ equity market.This systematic review found 21 types of biases affecting the investment decisions of investors.The study also found that there are few females included in the study who are unable to exhibit a true and fair picture of investors' behavior in the market.The studies identified that behavioral biases have a significant or insignificant effect on investment decisions depending upon the area of investment and population target.The cryptocurrency market is an emerging financial market, recent research is also found related to this context.

Future Research Directions
Further research can be carried out by researchers in various other contexts like Pension funds, life insurance, gold, bitcoins, etc. as there is a very less number of studies undertaken in these areas.It is also necessary to understand the investment behavior of investors investing in different avenues other than the stock market and mutual funds, researchers also need to carry out their research in the above-mentioned context.Future research can also be undertaken in comparing the investment behavior of investors investing in the same avenues but belonging to different countries of the same continent.Even researchers can also compare the investment behavior of investors residing in different continents but investing in the same avenues.This will help to understand the cultural differences, values, opinions, and knowledge of investors from different countries and continents.Further moderation and mediation effect can also be assessed by including other variables in the study and assessing the effect of moderation and mediation of the same variable among different biases.

Limitations
Despite the ample search across various databases, some important studies might be missed due to including only selected databases searching for specific terms having open access studies published in the English language.Additionally, some related studies might also be missed due to including only peer-reviewed articles.Further studies should be undertaken by researchers by broadening the search term and including more databases and studies published in other languages.

Figure 1 .
Figure 1.PRISMA flow diagram for the process of paper selection used in the systematic review study ofWolcott (1994)  and the action flows proposed byMiles and Huberman (1994)    Various behavioral factors affect ship investors while making their investment decisions.The study identifies the factors affecting the investment decisions of ship investors are ship finance, profile and business models of ship investors, market timing, and ship specifications.And all these factors of investment decisions are impacted by most behavioral biases.a strong association between availability and investment behavior of investors.As a strong relationship was found, Availability influences the investment decisions made by unit trust investors.
shows that the investment decisions of retail investors are significantly influenced by both herding and loss-aversion bias.The results also identify that the mediating role of Fear of missing out (FOMO) on the relationship of herding and loss-aversion with investment decisions amplified significantly.
this study reveal that overconfidence is present in investors' investment decisions in both markets.Overconfidence bias also influences the trading activities of individual investors at ETFs.

Table 1 .
. Information regarding Country, context, and types of biases

Table 2 .
Information regarding gender distribution, sample size, and target population

Table 3 .
Information regarding data collection, tools and results, and conclusion herding have a significant and positive effect on Egyptian investors' decisions.However, the gambler's fallacy does not affect significantly.Egyptian investors are overconfident, slightly regret-averse, and moderately loss-averse.Loss and regret aversion negatively affected the investment decisions of Egyptian investors.After COVID-19, herding behavior and gambler's fallacy do not affect the investment decisions of Egyptian investors.Researchers also conclude that behavior finance theory is valid before the pandemic and not valid after the pandemic in the Egyptian stock market.
The study found that all the biases prevail not only at the aggregate level of PSE but also in individual investors.The study also found that overconfidence is statistically significant as a mediator.