Chapter I

Introduction

Behavioral finance is defined by Shefrin (1999) as a rapidly growing area that deals with
the irdluence. of psychology on the behavior of financial practitioners. Behavioral finance
is a new paradigm of finance, which seeks to supplement the modem theories of finance by
introducing behavioral aspects to the decision making prcx;ess. It focuses on the application
of psychological and economic prinGiples for the improvement of financial decision
making as per Olsen (1998).

Similarly, Barberis and Thaler (2001) defined behavioral finance as a subject which studies
finance incorporating psychological behavior of the individuals and where all individuals
are riot always rational. As per Barberis and Thaler, the behavioral finance incorporaies
iwo building blocks. They arc limit to arbitrage and psychology. The limit to arbitrage
argues that if can be difficult for ihe mtional traders to undo the dislocations caused by less
rational trader and the aspect of psychology argues about the deviation occurring in the
individual behavior from the rationality due to emotional and cognitive values.

According tn Chitra and Jayshree (2014), Behavioral finance integrates individual behavior
and market phenomenon and attempts to identify the behavioral biases commonly
exhibited by investors and also provides strategies to overcome them.

Brabzoon (2000) studied the investor behavior in stock market and concluded that the
investor were not always rational and were influenced by behavinral biases. Their
decisions were also influenced by the behavioral bias.

This study primarily focuses on decision making process of investors, which is influenced
by the behavioral bias.
lJ Problem statement
Hasan, Khaiid and Habib (2014) identified the influence of two behavioral bias: loss
aversion and nverconfidence bias in the investment decision. The study also showed that
loss and overconfidence bias were influenced by age and gender.

In another study, Babajidi and Aditiloye (2012) studied other biases which affected the
financial decision making. The impact of overconfidence, loss aversion, training,
anchoring and status quo bias on Nigeria stock during last twenty years were outlined. In
the paper, it was also concluded that bias can be reduced through the assistance of

Chira, Adams, and Thornton (2008) targeted business students and studied how cognitive
and herustics affect the* decision making process. The study was conducted through
questionnaire and designed to check the behavior mistakes that they make during both
financial and non-financial decision making The data pointed on presence of bias and
heuristics on the students and these biases were overconfidencq excessive optimism, loss
aversion, familiarity, sunk cost, illusion of control and confirmation biases. Also the author
concluded that student rationality is bounded in their decision making behavior. The
business smdents were also not immune tn the biases.

Bashir, Javed, Ali, Meer and Naseem (2013) also identified impact of overconfidence,
cnnfirmation, illusion lnss aversion, mentai accounting, status quo and
excessive optimism bias in the financial decision making of Pakistani invesmrs.

Likewise, Adhikari (2010) studied the impact of herd behavior and overconfidence bias on
Nepalese investnr. Still there are many other biases to be tested among the Nepalese
investor. TNs leads to a research gap where different biases are yet tn be tested and aiso
tbeir influence in financial decisions.

The current study will focus on the overconfidence. confirmation. illusion of contml, loss
aversion, mental accounting, status quo and excessive optimism biases and their influence
on Nepalese investor’s financial decision making prncess. Similarly, the strenga of
relaoonship between these behavioral bias and Nepalese investors financial decision

2
making will be analyzed. Also the influence of demographic variable on the biases will be
examined. The biases held by individual investor will be compared with the biases held by
the institutional investors. The study deals with the following issues:

• What bias can be seen in Nepalese investor?

• What groups of people are more susceptible to the bias?
• Is there any relatinn between different biases and investment decisinn*
• On what extent is the bias associated with financial decision making?

ldObjectives

The main objective of the study is to idenbfy the biases on the Nepalese irrveatnr’s
financial decision making process. The other specific objectives nf the study are as
follOWS;

• To examine the relabonship between these behavioral biases and inveBtor’s

financial decision making.
• To analyze the relationship of demographic variable on these biases.
• To analyze biases held by individual and institutional investors.

1.4 Hypothesis

A total nf eight hypotheses was set for the study and tested as well. These hypotheses
were developed on the basis of past literature and empirical study.

HI: Loss aversion bins has effect on investors’ decision making process.

Odean (1998) concluded that individual investor demonstrated significant preference for
selling the winning investment while holding the losing one. Their decisions were
motivated by the fear of loss. The fear .of loss prevented the investor to trade ihe
investment.
H2: There is an impact of overconfidence on investors’ finam:ref decision making,

Chen, Kim, Nofsinger and Rui (2(D7) studied overconfidence bias in China. The srudy

bias along the other bias. They seem to believe that past reiums are the indicative of future

H3: Excessive optimism has effect on investors’ financial decision making.

Kafayat (2014) studied excess optimism bias in Islamabad Stock Exchange. The result
concluded that those who have excess optimism tendency make suboptimal decisions and
their outcome ih less then what they expect.

H4: There is no influence of mental accounting on inveabnent decisions.

Rockenbach (2004) reposed an experiment on the pricing of financial options. The

respondent had io evaluate different financial options- stock, bond and option. The
respondent associated the two risky assets (stock and option) to the same mental account,
while the bond is treated in a separate account. This shows the influence of mental
accounting on financial decisions.

HS: There is an inDuence of illusion of control on investors’ financial decision

Ajmal, Mufti and Shah (2012) studied illusion of control impact or stock trading behavior.
The study concluded that the illusion of control hns impact on perceived efficiency of
Pakistan investors. The perceived efficiency influenced the financial decisions.

H6: There is an effect of conBrmarinn bias on investment decision making

Onaomu (2014) identified the tendency to accept the favorable information and reject the
unfavorable information was prevailed in the individual investors. This was the case of
confirmauon bias. This tendency also affected ihe way they take investment decisions.
Samuelson and Zeckhauser (1998) conducted a series of derision making experiment
which showed that the individual stick with their initial decision i.e. maintain a status quo.
This particular bias played substantial role in the decision making of individual.

H8: Behavioral biases have impact on investors’ financial decision

The study of Adhikari (2010) on Nepalese investor concluded that Nepalese investor’s
financial decisions were also influenced by behavioral biases. Bashir, Javed. AI i, Meer and
Naseem (2013) also surveyed on Pakistani investor and i’ound the presence of behavioral
bias on Pakistani investor.

1.5 Definition of the terms

Standard finance is alsn known as modern pnrtfolin theory which has four foBndafion
blocks investors are rational; market.s are efficient; in ve.store should design their portfolios
according to the rules of mean-variance portfolio theory and expected return.s are a
function of risk and risk alone.

Bebaviozal I aaoce

Behavioral finance is an extension of standard finance which adds aspect of psychology in

standard finance. According to behavioral finance, investors are normal, not rational. A!so,
markets are not efficient, even if they are difficuli to beat. Investors design portfolios
according to the rules of behavioral portfolio theory, not mean-variance portfolio theory.

Individual investors are the individuals who managen or invest their own money in order io
achieve personal financial goals.

5
Institutioruil investor

Institutional investors are the. organizations which pool large sums of money and invest
those sums in securities, real property and other investment assets.

Prospect theory is a behavioral economic theory that describes the way people choose
beiween probabilistic alternatives that involve risk, where the probabilities of outcomes are
known.

Loss aversion

Loss aVerSion refers io people’s tendency to strongly prefer avoiding losses to acquiring
gains. Loss aversion recognizes mental pain and reward associated with loss and gain for
individual.

Overconfideoce

The overconfidence is a bias in which a person’s subjective confidence in his or her

judgments is reliably greater than the objective accuracy of those judgments, especially
when confidence is relatively high.

Confirmation

Confirmation bias, also called myside bias, is the tendency to search for, interpret, or recall
information in a way that confirms one's beliefs or hypotheses. li is a type of cognitive
bias and a systematic error of inductive reasoning.

Status quo bias is a cognitive bias; a preference for the current state of affairs. The current
baseline tor status quo) is taken +s n reference point, and any change frnm tbni baseline is
perceived as a loss.

6
Mental accounting

Mental accounting contends thai individuals divide their current and fuiure assets into
separate, non-transferable portions. The aeory purports individuals assign different levels
of utility io each asset group, which affects their consumption decisions and other
behaviors.

The illusion nf control is the tendency for people tn overestimate their ability tn control
events; for example, it occurs when someone feels a sense of control over outcomes but in
fact ihe individual cannot influence it.

The optimism bias is a bias that causes a person to believe that they are less at risk of
experiencing a negative event compared to others.

1.6 Scope and significance

The aim of study is to explain the influence of behavioral biases on the .investor financial
decision making. The past literature has showed that bias influences the financial decision
of the investors. The outcomes of the study will be helpful for the investors, policy nukers,
financial advisor, and students. The individual investor can take help fmm the findings of
this study and can come to know which bias interrupted their decision making, by
overcoming these biases they can make good investment decisions. Thus individual can
understand the pitfall and make their decision wiser, free from bias.

In the Nepalese stock market where information plnys a vital role, the findings of the study
can provide a framework to understand the behavioral aspeci of individual in the
investment decision making. The individual investor can be nware where were they going
wrong in the process of investment decision making and understand they way institutional
investor make decisi an.
Likewise the study has some contribution to the existing literature. The contribution of this
study in the existing literature is that aere was no much study where all these biases i,e
status quo, overconfidence, confirmation bias, illusion of control, excessive optimism, loss
of aversion, and mental accounting have been discussed together.

1.7 Limitations

The study was limited by various factors which are mentioned below:

• The data collection for the study was conducted before and after the mega
eanhquake that Nt Nepal in 2015. There was psychological impact of
earthquake and its continuous afiershocks on the respondents. The
psychology of respondent during that time period could have impaci on the
response they provided.
• The respondent of the study were ihe investor who were. residing in
Kathmandu Valley. Therefore, the study is limited to re.spondents residing
in Kathmandu Valley only.
• The study of behaViorai bias nn iissbtutional investor was carried out by
conducting four personal interviews with which may not be sufficient to
generalize the study result.
• Pompian (2006) has pointed twenty behavioral biases while the study
covers only seven out of those twenty behavioral biases.

1.B Outline of the report

This GRP repori consist of three main sections: preliminary report, body and
supplementary. The preliminary part includes title page, certification, declaration of
authenticity, acknowledgements, table of contents, list of table; list of figures, common
abbreviations used in the report and executive summary.

8
The body of ihe report, which is the. main part, includes five sections: introduction,
literanirc review, research methods, data analysis and results and discussion, conclusion
and implications.

Chapter 1: Introduction

The introduction action of the report includes the background informañon on the
relationship between investment decision and investor behavior. The relationship has been
derived on the basis of past literature. Apart from this, this section mainly consist problem
statement, objecfives of study, hypothesis, limitaoon and scope and significance of the
study

In the second chapter, the literature review related to traditional finance, behavioral
finance, prospect theory, behavioral biases to be tested in the study and theoretical
framework bas been presented. The literature has been presented separately for each
section. The chapter further consists of theoretical framework which explains the
relationship between the dependent and the independent variable with the presence of
moderating variable.

CMper3:RmearrbmetWdslogy

The third chapter consists of ihe outline of methodology used for the study. It includes
descriptions about the research design, sample for study, nature and sources of data,
in6truraents and]2rocedures used for the study and a brief introduction to the data analysis.

Chapter 4: Data analysis and results

The fourth chapter- data analysis and results consists the analysis and interpretation of data
using quantitative and qualitative methods with the assistance of statistical and
mathematical modrls. The SPSS and MS- Excel software has been used for the data
analysis and interpretation has been done accordingly.

9
Chapter 5: Discussion, conclusion and implications

The filth chapter deals with discussion, conclusion and implicaiion of the study. The
discussion part includes ae integration of theory with tisat of the findings of ihe study,
investigation of the study and drawing out of conclusion and implications.

The final Aerion of the repori consists of supplementary section. This section consists of
references and appendix as questionnaire and summary of responses. The references are
presented in APA format.

10
 Chapter II

Related Literature and Theoretical Framework

This section compromise of the literature carried out to carry out ihe study and theoretical
framewo& which defines dependent, independent and moderating variable.

2.1 Standard finance

As per Ricciardi and Simon (2000), standard finance theories are associated with the
modem portfolio theory (MPT) and efficient market hypothesis (EMH). Similarly, Siatman
(1999) further added the other pillars of standard finance as arbitrage principle of
Modigliani and Miller (1958), capital asset pricing theory of Sharpe (1964) and option
pricing theory of Black, Scholes and Merton (1973).

Markowitz’s theory is today known as the modem portfolio theory, (MPT). As per
Markowitz (1952), MPT is a theory of investment which attempts to maximize portfolio
expected return for a given amount of portfolio risk, or evenly minimize risk for a given
level of expected return, by chnosing the proportinns pf varinus assets.

Likewise another pillar of standard finance is efficient market hypothesis (EMH). The
EMH theory was formulated by Fama (1970 & 1991). Fama (1970) defined an efficient
market as a market in which prices always fully reflect all information. It is the proposition
that current stock priGes fully reflect available information aboui the value of the firm, and
there is no way to earn exceas prnfits by uaing this information. The EMH was supported
by Malkiel (2003), who examined the attacks on the EHM and concluded that steck
markets are far more efficient and far less predictable.

EMH have three basic foundations: rational traders, arbitrage opportunity and subjective
utility function. Thus EMH have limitation or criticism on the basis of these assumptions.
Kahneman and Riepe (1998) concluded that investors deviate from economic rationality
which criticizes the assumption of rational trader. Also, Mitchell, Pulvino, and Stafford
(20ti2) were able to document 82 cases in which the market value of a company is less than
the market value of the company’s stake in its subsidiary. There was presence of arbitrage

11
opportunity but the authors found a degree of persistence that indicated barriers to
arbitrage.

2.2Behavioral finance

Behavioral finance is a branch of finance thnt studies the behavior of individual in the
financial market which are influenced by psychological. Behavioral finance aims to
explain the reasons behind the irrationality of investor and efficiency of market.

Kahneman and Tversky ( t979) have presented behavioral finance as a theory which adds
the aspect of psychology and economics in the traditional finance theory. Likewise they
have presented critique for expected utility theory and proposed prospect theory as an
alternative which aiso gives base for behavioral biases.

Similarly, Ricciardi and Simmon (2000) presented behavioral finance as multidisciplinary

subject compromising of finance, psychology and sociology. Shefrin t1999) defined
behavioral finance as a rapidly growing area that deals with the influence of psychology on
the behavior of financial practitioners.

According to Ritter (2003), behavioral finance has two building blocks: cognitive
psychology and limits to arbitrage. Under the cognitive biases, author lists heuristics, over-
confidence, mental accounting, framing, representativeness, conservatism, and disposition
effect. Ail these psychological facton are related io financial markets which explains why
they do not behave as traditional financial theory requires. The limit to arbitrage is the
second building block of behavioral finance. Author pointed that arbitrage is possible in
shori-ierm frequently recurring aituaiioris, but long-term mispricing are corrected and
arbitrage is hence limited.

Behavinrai finance can be segregated intn two main segments i.e behavinral finance micrn
and behavioral finance macro. Behavioral finance micro (BFMI) examine.s behaviors or
biases of individual investore. Whi g:, Behavioral finance macro (BFMA) detects and
describe anomalies in the efficient market hypothesis that behavioral models may explain.
(Pompian, 2006)

12
As per Kiyilar and Acer (20D9) Macro-behavioral finance covers the investment behaviors
diat strive to explain the anomalies on the effective market hypothesis, Micro-behavioral
finance strives to explain the individual investors who differentiate themselves from the
rational investors indicated by the classical economical theory.

2N Prospect theory

Kahneman and Tversky (1979) proposed prospect theory. The theory explains the way
individual choose between probabilistic alternatives that involve risk. In other words, it
explains how individual manage risk under uncertainty. The theory states thai individual
make decisions based on the potential value nf losses and gains rather than the final
outcome, and that individual evaluate these losses and gains using cenain heuristics.

According to the theory, individual place more weight on the certain events in compared io
the probable one. The model arse describes the framing effect. Likewise, the theory further
explains thai individual value gain and loss differently. individual will feel higher degree
of pain in case of loss and lower degree of joy in case of gain even if the gain and loss
amount remains the same.

The scenario where one individual gains $50 should be equal the scenario where the
individual gains $100 followed by loss of $50. In both the scenario the nei amount at the
end is $50. Hnwever the individual will take the first scenario mnrr favorably than the
second one.

The r gz .‹hows the difference in utility. Utility is defined as amount of joy or pain for the
individual. Figure 2.1 is the value function for the individual. Figure shows the scenario
where individiial have Sâ0 gain and $50 loss frnm a common reference point. The amount
of gain and loss is some i.e $50. The figure explains that the degree of joy is smaller than
the degree of pain although the amount is same.
 Flgure Al Prospers theory

When there are multiple cases of gain and loss, each event is valued separately and then
combined as one. According to value function, gaining $50 will cause +10 points of joy
and loosing $50 will cause -50 points of pain. These points are measured in utility. If a
person gains $50 and soon lose the same amount then the net payoff will be -40 units .of
utility. This shows that there is huge pain when there it equal amount of gain and loss.

Prospect theory also explains ihe disposition effect. Disposifion effect is the tendency of
the individual to hold the loosing investment for a long period of time and get rid of the
Wi£i£Ii£I@ irIV9.StIN0rTt £00 0oOrt.

2.4 Loss aversion

The principle of loss aversion was first introduced by Kahneman and Tversky (1979).
According to Kahneman and Tversky (1979), loss aversion is related to individual stronger
desire to avoid losses than comparable gains. According to the value function of prospect
theory, gains lead to risk aversion, while losses lead to risk seeking, Kahneman and
Tversky pointed further on loss aversion. People are not reañy to accept losses and u'ill

14
eventually take higher risk to try to mitigate facing sure losses. If trading by an invtsior
follows the. disposition efféct, it implies thai the investor’s decision to sell a steck depends
on the purchase price. However, the price paid for the atock is a sunk cost at the time of
sale, and abseni tax considerations should not affect the timing of sale of a rational
investor.

Barberis and Huang also supported Kabneinan and Tversky concept of loss
aversion and explained that behavioral finance assumes investors as loss averse than risk
averse. They incorporeted the loss aversion into utility function and found investor wilt
bear higher degree of pain in loss while comparing with the degree of joy if the invntor
have gained same ainoum. Thus, investors will increase their risk, defined in ierms of
uncertainty to avoid even the smallest pmbability of loss. An example of an assumption
about preferences is that people are loss averse: a $2 gain might make people feel bener by
as much as a $1 loss makes them feel worse. They further showed thai loss aversion in
individual stocks leads to excess stock price fluctuations.

Odean (1998) added another dimension i,e, decision making behavior of investnr to study
the influence of loss aversion on financial decisions and concluded that there exists
tendency of investor to sell the u'inning investment and bold the losing one for a long
period of time. for the study the trading account of investor were studied. The strong
preference to hold the winning investment was identified.

Covai and Shumway (20fl5) carried out similar study as of Odean (1998). Their study
showed that traders on the Chicago Board of trade exchange take more risk late in the day
tn cover their losses in the beginning of the day. This implies loss aversion behavior. Prices
are affected by this behavior in that they are willing to buy contracts at higher prices and
vice versa than those that prevailed earlier.

In both study of Odcan and Coval and Sbumway, loss aversion affected the investment
decisions or influences the behavior of individual. A slightly different study was conducted
by .Gachter, Jhonson and Herrman (2007) where they tried to see influence of loss aversion
between risky and riskless investment. There was no evidence to support whether people
who are loss averse in riskless choices are also loss averse in risky choices. Loss aversion

15
in the riskless choice task and loss aversion in the risky choice task are highly significantly
and siroogly positively correlated.

2JExcessopdmien

People are overoptimistic when they believe that they will not be exposed to future events
and things will not go beyond their oonirol and people also think that there is high
probability that positive event will happen to them and negative events will happen to
others (Weinstein, 1980). In further explanation, author asserted that people have tendency
to believe that it is less likely that they will suffer from an undesired situation then the
others. The author also studied ihat excess optimism bias is in many domains and age

One of the study in excess opt mism was carried by (Brown & Cliff, 2005) who in their
study pointed that the investor feelings do influence their ability to value a security. Due to
their optimistic nature, they vaiue security highly. The aspeft of decision making was
6tfldied in different study. Two of those studies were conducted by Sadi, Asl and Rostami
(2011) and Kafayat (2014). In both the study, the influence of excess optimism ovo
decision making was studied among the individual investor.

Sadi, Asl and Rostami (201 l) analysed the impaci of perceptual bias and personality on
financial decision making process. The perceptual biases which included handshight,
overconfidenm, optimism and avaiIebility bias influenced the decision making process.
The investms who were influenced by excess optimism showed different decision taking
behavior in compared to the investor with low or average level of optimism.

sim:iariy. Kafayat (2014) aim studied impact of excess opfimism bias in Islamabad Stock
Exchange. The result concluded that those who have excess optimism tendency make
suboptimal decisions and tbeir outcome is less then what they expect. Furthermore the
evidence showed that the investors who were not exposed or aware of the biases make
rational decisions and thus they enjoy more favorable outcome.

16
2.6 Overconfldence

.odean (1998) finds that investors tend to overestimate their ability, unrealistically
optimistic about future events, too positive on self-evaluations, over-weight attention
getting information that is consistent with their existing beliefs, and over-estimate the
precision of their own private infomiation.

Odean conducted two study with Barber in order to examine the impact. of overconfidence
bias on investor behavior. In their study of overconfidence bias, (Barber and Odean, 2001)
found that it effects the rational decision making of investon and wilt lead the investor to
trade in more risky security. Similarly, BarLer and Odean (2000) also studied that when
investor are overconfident, they trade too much. In both of the study, it was concluded that
overconfidence has influence in the investor.

The output from the study of Barber and Odean (2000 and 2tXll) was supported by Chen,
Kim, Nofsin r and Rul (2007) who studied overconfidence and representative bias in
China. The study concluded that Chinese investors make poor trading decisions impacted
by behavioral biases. They tend io sell stocks that have appreciated in prices, but not those
that have depreciated, they seem to be overconfident, and they seem to believe that past
returns are the indicative of fiiture returns. The dataset caine fmm a brokerage firm of
SHSE and SZSE in China. The complete dataset included individual investor and
institutional investors. Regrmsion analysis concluded that Chinese investors make trading
mistakes, they are reluctant to realize their losses and they tend to be overconfident.

More study was conducted on nverconfidence bias which indicated the excess trade was
dune by overoonfident investor. As per Tapia and Yermo (2000), People generally rate
themselves as being above average in their abilities. They also overestimate the precision
of their knowledge and their .knowledge relative to others. Many investors believe they can
beat the market. But in reality there's an overwhelming amount of evidence thai proves
otherwise. Overconfidence resiilrs. in excess trades, with trading costs lower return.

17
2.7 Mental accounting

Thaler (1999) described mental accounting as whereby people code, categorize and
evaluate economic outcomes. As per Thaler, concept of framing is important in mental
accounong analysis. The concept of framing was presemed in his previous study. Thaler
(1985) performed an experiment in which he offered. one group of people $30 and an
choice: either pick the money or gamble on a coin loss, where a min would add $9 and
a loss would subtract $9 from the initial $30. Seventy percent of the people offered this
choice elected to gamble, becaum they considered the $30 io H endowed money. A
second group of people were offered a slightly different choice. Outright, they were
given choice if they will gamble on a coin toss, in which they will receive $39 for a win
and $21 for a loss or would you raiher simply pocket $30 and leave coin toss gamble.
The key difference was that the the second group were not offered $30 outright as in the
first group were offered. The options were presented in terms of their ultimate payoffs.
The second group reacted differently from the first Only 34 percent of them chose to
gamble, even though the economic prospects they faced were idcntiGal to those offered
io group one. Sometimes people create mentai accounts in order to justify actions that
seem enticing but that are, in fact, unwiw. Other times, people derive benefits from
menial accounting.

Mental accounting impacts the investor’s way of making financial decisions,

Rockenbach (2HJ4) carried an experiment on the pricing of financial options. The data
showed thai even with considerable experience, unexploited arbitrage oppominities
persist. Subjects did nor seem to make the connections bniween the different investment
possibilifies, as

risky ansets (stock and option) to the aune mental account, u'hile the bond was treated in a
separate account.

But. in the study of Bamndagh and Hasanzfldeh (2013) il was concluded iJi«i there was
no impact of mental accounting in invesnnent decisions. The study mcthnd was
descriptive — correlative. Findings showed that there was a negative relationship
between accounting and decisions of investors, assessments of financial activities, and
assessments of financial dictions. And also, there was no significant relationship
between mental accounting and resource allocations.
These two studies produced contradicting result in case of mental accounting. These were
in case of individual investor. The concept of Mental Accounting bias was prevailed in
case of insotutional investors also. Fisher and Statman (1997) reported that mwtual fund
companies often recommend constructing ponfolios as pyramids of assets with cash in the
bouom layer, bonds in the middle layer, and stocks in the top layer.

2.8 Illusion of cmtrol

Langer (1975) defined the illusion of control as an expectancy of a personal success

probability inappropriately higher than the objective probability would warrant Langer
found that choice, task familiarity, competition, and active involvement ali lead to inflated
confidence. beliefs. For example, danger found that people who were permitted to select
their own numbers in a lottery game demanded a higher price fur their ticket than did
people who were assigned random numbers.

Moore, Kurtzberg, Fox and Bazemian (1999) examined portfolio allocation decision of
business students. The computer simulation method was used and the study aimed tn study
why people incest on managed portfolio much. They concluded thai investment decisions
were susceptible to positive illiuions and overestimation of inter temporal consistency.

Another study was also conhuced among the students. Breinholt and Dalyrmple (2004)
examined the Hlusinn of control nn undergraduate students. The illusinn nf control was
smdied as interaction of the desire for control and the belief in good luck. The
susceptibility to illusion of control was determined by these faciors. Participants were
randomly assigned to either a high involvement or low involvement condition and either a
descending or random sequence of outcomes. Results showed thai under low involvement
conditions one's belief in luck will predict amount wagered whereas under high
involvement conditions desire for control will reliably predict amount wagered. This study
clearly demonstrated the iPuéon of control bias in practice.

.As on studenn, the study was conducted among the investor. Creevy, Nicholson, Soane, &
Willman (2fD3) examined the impact of illusion of control on the performance of traders in

19
financial instruments. They argued that tendency to illusions of control bias would be
related to trader’s performance. The study revealed that traders with higher level of illusion
of control performed less as compared with those with lower level.

The previous study was supported by Ajmal, Mufii and Shah (2012) and concluded that
Illusion of control impact on stock trading behavior. It was outlined that Hlusion of control
directly or indirectly has an impact on trading behavior in the stock market, i.e.
overreaction and under reaction of stocks.

2.9 ConDrmatiort

Wickens and Hollands (2HJ0) defined the confimiation bias as a tendency for people to
seek information and cues that confirm the tentatively held hypothesis or belief, and not
seek for discount) those that support an opposite conclusion or belief.

Bogan and Just (2009) studied confirmation bias in the behavior of actual corporate
executives. The experimental study was conducted and frequency technique was used for
the study purpose. The study included observations from students and higher executives.
This study concluded that higher executives were less likely to absorb the new informaoon
in contrast to non-executives.

Unlike Bogan and Just, Rachel (2007) investigated the confirmatory bias behavior in
investor. Data were cnllected from participants thrnugh confidenoal survey in the United
States. The descriptive state and frequency distribution technique was used to investigate
Ifte confinnatory bias behavior aztthio ittveetot’s decision making. The etudy canaluded
that the confirmatory bias affected the evidence related losses strongly as compared to
evidence related gains.

Onsomu (2014) also identified behavioral biases which aHect individual investors at the
Nairobi securities exchange. fn addioon, the relationship between gender and the
behavioral biases was investigated. The results indicated thai investnrs are affected by
confimiation bias. fiemale were more affected by the confirmation bias.

20
2.10Status quo

Samuelson and Zeckhauser (1998) introduced the status quo bias as an emotional bias that
gives people facing a number of choice options to choose whatever option ratifies or
extends the existing condition in lieu of alternative options that might bring about change.
They conducted a series of decision making experiment which showed status quo bias on
individuals. The smdy concluded that the individual showed stams quo alternatives that is
doing nothing or maintaining the current level. A series of question were asked to the
respondents which show«t the siatus quo bias in those respondents.

Madman and Shea (2000) also found that investors are subject to statun quo bias and tend
tn stick to their initial decisions in their investment decisions. Likewise, Kent, HirsMeifer
and Teoh (2001) found that investors gave limited attention and processing power io their
decision making. This ifi due to their stauis quo since ihey interpreted that the status quo
option is an implicit recommendation.

Also, Yahyazadehfar, Shams, Forughinia, & Alimardan (2012) observed Status quo bias in
their study. People choose previous or past path to reach a decision. The Ruenzi and
Kempf model was used. The data was collected in Teheran from investment related
companies. It was concluded that people exhibit. status quo bias showing preference for
alternative path related to their past experience.

AH these study showed that status quo bias influences the investment decision and investor
tends to stick in their initial status.

2.11Inflnence of behavioral bias In financial tension making

Kahneman and Riepe (1998) presented a paper which highlighted about behavioral bins
affecting the decisions. They termed behavioral bias and wrong judgment as cognitive
illusions which affect the decision making potential of the investors. They have focused on
eight different biases, out of which overconfidenoe and optimism were two. Likewise they
have also presented recommendations for financial advisor to overcome these biases.

21
Likewise, Riccardi and Baker (2014) mentioned that Investor displays various behavior
which affects the decision making process. Investors deviates from rationality and logics
and be influenced by the bias. The author have mentioned various biases and said that
although the bias cannot be completely avoided but can be managed but still the invesior
wifi be influenced by biaa. They further explained that investor exhibit status quo behavior
or overconfident behavior.

Also, Ashta (2005) claimed that behavioral bias affects the decision making in risky
project evaluation. He forwarded loss aversion, fronting, herd behavior, optimism and
anchoring as the biases which affected the decision making for risky projects. He has
forwarded the ale of psychology as base for these biases. The smdy was done in corporate
investment decision.

Suresh (2013) adopted exploratory and descriptive based on the figures from the
secondary data of the various financial reprint and speeches of financial advisors and
primary data from researcher’s and various financial expert’s ideas and opinions
mentioned that the understanding of behavioral biases can help individual take sound
financial decisions and in turn make him a better investor.

Adhikari (2010) surveyed on Nepalese investor and concluded that Nepalese investor’s
financial decisions are also influenced by behavioral biases. These behavioral bias were
herding behavior and overconfidence.

Likewise, Bashir, Javed, Ali, Meer and Naseem (2013) also surveyed on Pakistani investor
and found the premce of behavioral bia6 on Pakistani investor. They too found similar
result as of Adhikari (2010). The behavioral bias affected the decision making process of
Pakistani investor as of Nepalese investor.

22
2.12 Theoretical framework
Independent variable

Overconfidence

Excessive optimism

Mental accounting Nepalese investor’s

" financial decisions

Confirmation

Status quo

Moderating variable

Age
Gender
Education
Occupation

23
2.12.1 Dependent variable

According to Sekaran and Bougie (2012) dependent variable is the one on which the
researcher have primary interest. The researcher tries to predict, describe or explain its
variability. The variability comes from the change in those variables on whicb it depends.
financial decision making is the dependent variable.

Decision-making can be defined as the process of choosing a particular alternative from a

number of alternatives. It is an activity thai follows after proper evaluation of all the
alternatives. (Mathewa, 20fi5).

2.12a Independent variable.

As per Sekaran and Bougie (2012) independent variable is the one which influences the
dependent variable in either positive or ill negadve way. In this case, behavioral bias
affects the decision making so behavioral biases are independent variable.

loss aversion refers to the tendency for people to strongly prefer avoiding lossn than
acquiring gains, Loss aversion we nest convincingly demonstrai‹›d by Tversky and
Kahneman- Pompian (2006).

Overcanfideace

The .overconfidence is a bias in which a people overestimate both their own predictive
abilibea and the precision of the informabon iheyM been given. It is a tendency in which
a person. subjective confidence in his or her judgments is reliably greater than the
objective accuracy of those judgments, especially wten confidence is relatively high -
Pompian (2006).

Confirmation bias, also called myside bias, is the tendency a have type of selective
perception that emphasize ideas that confirm our beliefs, while devaluing whatever

24
contradicts our beliefs. II is a type nf cognitive bias and a systematic error of inductive
reasoning -Pompia n t2006).

Status quo bias is an emotional bias that predisposes people facing an array of choice
options io elect whatever opiion ratifies or extends the existing condition (i.e. status quo)
in lieu of alternative options that might bring about change -Pompian (2%6).

Mental accounting

Mental accounting contends that individuals divide their current and fuiure assets into
separate, non-transferable portinris. The theory purports individuals assign diHerent levels
of utility to each assei group, affects their consumption decisions and other
behaviors -Pompian (200fi).

Tire illusion of control is the tendency for people to overestimate their ability to control
events; fur example, it ocfurs when someone feels a sense of control over outcomes that
they demonntrably do not influence- Pompian (2006).

Excess optimism

The opomism bias is a bias that causes a person tn beheve that they are less at risk of
experiencing a negative event compared to others- Pompian (2006).

2.12 J ModeraUag variable

Sekaran and Bougie (2012) has again defined moderating variables as the one which has a
strong contigent effect on the independent variable — dependent variable relationship. fn
this case age, gender, occupation and education background are moderaung variable.

25
2.13 Concluding remarks
From the previous study, it can be concluded that behavioral finance has slight edge over
the standard finance. Investors are not always rational and not always logical beiund their
decisions. There will be influence of psychological and behavioral factor arid hence
investor will exhibit certain bias. There biases will impact the financial decisions. However,
much more study is yet to be done in order to under.stand more depth regarding behavioral
bias and financial decisions.

26
 Chapter III

Research Methodology
The chaptef is mainly concerned with the procedures that have been used tn collect and
analyze the data for the study. The chapter deals how the study was carried out in order to
test the hypothesis and fulfill. the objectives set for the study.

3.1 Research design

Malhotra and Birks (2000) have described research design as overall plan for connecong
the conceptual research problems to the empirical rmearch. In other words, the research
design a ticulaies what data is required, what methods are going to be used In collier and
analyze this data, and how art of this is going to answer your research question.

The study desiga for the study was basically descriptive and aimed at describing the state
of affair, as it exists at present. As per Sekaran and Bougie (2012) a descriptive study is
undertaken in order to ascertain and be able to describe the characteñstics of variable of
interest. The basic methodology used in the study was quesnonnaire survey.

3J Population and sample size

This study aimed to collect data from Nepalese investor and the data was collected from
the rmpondeno who were presem at stock brnkemgc house. There is no clear data
regarding the investor size in NEPSE.

The minimum sample sine was 120 for individual investor. Roscee (1975) proposed the
nile of thumb for unknown population. When sample has io be broken inio subsamples
tfien there must be minimum of 30 samples in each sub category. Rnscoe further explained
that in order in derive a valid resuli, there must be a minimum of 30 responses in ench sub
category of parameters.

There were four categorical variables in questionnaire where sample can be broken down
into subsample and in one variable there were four sub categories for subsample. So this

27
justiEed the minimum sample size of 120 for individual investors. However, data from 136
respondents was collected to fulfill the minimum number of 30 in each category mle.

Likewise, four institutional investors were selected for interview. The institutional
investors were selected from the mutual fund listed in NEPSB. There were five motuai
fxind listed in NEPSE. But the number of organization to issue mutual fund were only four
as Siddharth Capital has issued 2 mutual fund. Out of these mutual fund three organization
were to be selected for interview. Raosoft Sample size Calculator (2OM) concluded that
with population of rve the sample size can be three if the response rate is on higher side.
However, due to lack of response for interview, one insurance company was selected as a
replacement.

3d Instruments

In order to draw data to fulfill the objectives, 9uestionnaire and interview method as
instruinenis were used in the study.

The questionnaire for individual investor contained 17 questions out of which 7 question
were related to decision making in Likert scale form and other remaining 10 question were
used to test the bias of the respondents. The questions used to test the bias were single
response questions. The questionnaire was in English and Nepali for the
simplicity of the investors. The questionnaire was divided into three sections, The first
section included general introduction regarding the smdy and questionnaire, the semnd
secuon included the 9Uestion required for the study and the last section included the
demographic variable (age, gender, educational background and occupation).

Likewise, for institufional investors, in depth interview was carried out. A separate check
list was prepared and responses from each interview were noted down. The responses from
all interviews were merged, summarized and presented. The manager level employees of
institutional investor were the subject of interview.

28
3A Source of data

In this study both primary data were used. Primary data were obtained through
questionnaire for the individual investors and in depth interview for institutional
investors.

#JSOñW8IRUS

The data obtained fom the questionnaire response were analyzed using SPSS sofiware
and results thus obtained was analyzed, imerpreted and was presentt4 in the wrinen
formai

3.6 Data analysis

This kind of analysis provides a frequency tables that report the percentage of each of
categories and diagram that easy to interpret and understand. All of number and
percentage are produced with SPSS for windows. The following techniques were used:

• Pie Charts

• Cmss Tabulation

• Chi Square
• Correlation

• Cronbach Alpha

• Regression Analysis

3.7 Functional relationship

The functional relationship is defined on the basis of theoretical framework, where

financial decision moking is dependent variable while behavioral biases are independent
variable.

Z9
Y= Nepalese Investor Financial Decision Making
X l = Loss aversion
X2= Illusion of control
X3= Optimism
X4= Overconfidence

X5=Confirmatinn

X6= Mental accounting

X7- Status quo

e = Standard error

30
 Chapter IV

Analysis and Results

This section of the study presents the analysis of the daia collected. The analysis aims io
fulfill the objective of the study. Tlse analyses of data from individual and institutional
investor are presented in this section.

The chapter coniai as the analysis of demographic and other characteristics of individual
customer, behavioral bias with aspect to gender. age, educational background, occupation
and other respondent characteristics, reliability test using cronbach alpha, level of
influence in decision making, relationship between the behavioral biases, regression
analysis and quuJitdtive data analysis of institutional investor.

4.1 Respondents demographic profile

This section explains the demographic characteristics of the respondent. In this sectinn, the
respondents profile has been analyzed in terms of gender, age, educational background and

4.1.1 Respondents gender profile

In the study, data from 136 respondents has been collected. Figure 4.1 shows the
distributinn pf respondents on the basis of the gender. Out of 136 respondents, majority
(69.90&r) of the respondents are male respondents while remaining (31. ION) of the
respondents are female respondents.

30.1 G

Figure 4.1 Geoder wise disMbutions

31
4.IN Respondents age profile

Similarly, the figure 4.2 illusirates the distribution of respondent.s on the basis of age. On
the basis of age, most of the respondents fall in the age category of 20-30, which accounts
for 41.2 percent of the respondents and is followed by the age category of 40 and above
and of 30-40. Out of the remaining proportion, 30.9 percent of the respondents are above
the age of 40 and 27.9 percent of the respondent are in the age category of 30fi0.

I]gure€ZAgevdsed buéo-

4.IN Respondents education profile

The respondents were asked if they were from financial academic background or if they
were from non finance academic background. Majority of the respondent were from
finance background in stock market. In the figure 4.3, it can be seen that 57.4 percent of
the respondents were from financial background while 42.S percent of the respondent
belonged to non finance background.

Pigure 4J Education wlse d¥strlbutlons

32
4.lA Respondents occupation profile

Similarly in terms of occupation, the resp‹indents are almo.st equally distributed in ihe
occupation category except in business where the proportions of respondent are on lower
side. Out of 136, 29.40 percent of the respondents are students, 26.50 percent of the
respondents are investment tmder, 23.50 percent of the respondents are servicemen and
remaining 20.60 percent of the respondents are business person.

Investmen
t Trader, Service, g Se p i
26.50 23.50%
w Business

o Student

Business,
Investment
Student, 20.6V
Trader

Flgure 4.4 Occiipulon wlsz diiarlhiitlrias

4.2 Respondents risk, experience and trading frequency profile

rhe questionnaire was also designed to analyze the trading frequency, experience and risk
profile of respondents and influence of these vai iables on the behavioral bias.

Table 4.1

Respondriits ri.sk, experience and truJing fr«quen‹ y profile

Experience Basis
TLlsk Pzoffle Regularity ln Trading
0-2 year 45.6^<
Risk A ?8.2& Irrc#olar 66.2&
2-4 year 20.6&
acne Rielc 42.6&
4-h year 7.4
Neutral Risk 9. Ik Regular
6 year + 26J@
Cover

Table 4.1 explains the distribution of respondent on the basis of experience in siock
investment. risk profile and regularity in trading stock. The collected data shows that
nearly 65 percent of the respondent have less than 4 year of experience and remaining 35

33
percent of the respondent have more than 4 year nf experience. Likewise, On the basis of
risk profile, very few (19 IN) arc risk lover while other remaining respondent are either
risk averse or risk neutral. Likewise, 66.2 percent of the respondents are irregular in
trading. They dn not trade frequently. The other 33.8 percent of the respondent trade
regularly.

4J Behavioral bias in Nepalese stock investor

Table 4.2 shows that a large proportion of the investors are influenced by overconfidence
bias. This means that the investors are overconfident in their decisions. Alongside being
overconfiderit, they are arse optimiet and loss averse. Amund two third of the respondent
are influenced by these bias, Likewise, two third of the respondent were infiuenmd by the
menial accounting bias.

Table 4.2
Behavioral bias cusceptibility
Behavior Blue Suscaptlble (h) hmu**pBMe(W)
Loss Aversion 73.5 26.5
Ovcrconfidcnoc 77.9 22. i
Optimiun 75.0 25.U
Mentgl Accounting 75.0 25.0
Elusion of Control 25.0 75.0
Confiritntion 30.9 69. I
Status Quo 57.5 42.6

In case of other biases, the susceptibility was low. .Out of 136 respondents, only one fourth
were susceptible from illusion of control bias and nne diird for confirmation bias. 57.5
percent of the respondents are influenced by starus quo bias. They try to maintain their
initial status.

34
4A Behavioral biases with respect to respondent's background

This secoon explains the influence of behavioral bias with respect to the demographic
variables. The demographic vanabk includes gender, age, educational background and
occupation. The influence of these variables on the behavioral bias has been shown in the
tables below. The test of significance has heen done at 5 percent significance level.

4.4.1Behavioral Bius witb respect to geader

Table 4.3 shows the influence of gender on loss aversion bias. As per the data, both male
and female are almost equally susceptible to the loss aversion bias. 74.7 percent of the
male investor and 70.7 percent of the female investor have loss aversa tendency. However,
male investors are more loss averse in compared In female. Also by, comparing n, which is
6et at 5 percent and p-value, p-value is greater than ti so it can H concluded ihat there is no
significant relationship between gender and loss aversion bias.

Table 4.3

Loss aversion bias wifh respect io gender

Loss Aversion
Pearson Chi Squnre Asym. Sig (2- tniledJ
Susceptible (&) Insuseeptible {&)
Male 74.7& 25.3&
Ferrvde 70:7& 29.39b 0.236 0.627

Similarly, Table 4.4 reporis the influence of gender nn overconfidence bias. Male investors
are mnre overconfident compared to female. The .difference between level of
overconfidence between male investor and female investor is very less. The collected data
shows that 78.9 percent of tbe male investor and 75.6 percent of the female investor are
influenced by the overmnfidence bias.

35
Table 4.4

Overconfidence bins with respect in gender

Overconfidence Pearson Chi Asym Fig (2-
Gender
Susceptible (&) lnsuscepñble Square tailed)
Male
(%) 78.W 21.1 &
75.6W 24.4
0.187 0.667

The table also shows that there is rio significant relationship between gender and
overconfidence bias, where p-value is greater aan o value at 5 percent.
The Table 4.5 shows the influence of gender on optimism bits. The tabulaied data shows
that male investors (76.8& of male investor) are more optimists in compared to the female
investors (70.7& of the female in vestor). The table further shows thai there exists rio
significant relationship between gender and optimism bias.

Table 4.5
Optimism hias with re.syect to gender
Optimism
Pearson Chi Asyei, 5ig (2-
Gender
Suméptible (&) lnsuscepiible (&) Square tailed)
Male
76.8& 23.2&
hcmalc
70.7& 29.3&. 0.570 0,4H

Likewise Table 4.6 illustrates the impact of gender on mental accounting bias. In case of
mental accounting bias, male investor are more susceptible in compared to female investor
by a s ignificant margin. Comparing male and female investor, 80 percent of male
poswssed this bias while only 63.4 percent of female possessed menial accounting bias.
Similarly the table also shows that there is no significant relationship between gender and
overconfidence bias.

36
Table 4.6

Mental accounnng bias with respect to gender

Mental Accounting Pearson Chi Asym Fig (2-
Gender Susceptible (&) lnsuscepñble (%) Square tailed)
Male 8W 20%
63-4& 36.6 4.202 0.40

The data collected revealed different reset for illusion of contml bias as compared with other
bias tested for the study. Aimnst similar prnportion of male investors and femaie investors are
suscnptible of illusion of control Also, ihe rest shows that there exists no significant
relationship between gender and illusion of control..

Table 4.7

Illusion of control biac with respect to gender

illusion of Conlzol
Pearson Chi Asym, 5ig (2-
Gender Susceptible (%) !nsuxepiibJe (g) Square tailed)
Mae 23.396 74.796
Pemale 24.4 75.6& 0.12 8.914

Female investors seem to be more susmptible in compared to the male investors for
confimiation bias. Our of ail female respondents, 34.1 percent of the respondents have
shown the bias while 29.5 percent of ae male were only influenced by this particular bias.
The ie.st of significant relationship was similar with other pmvious leer. In this case
also there is no significant relationship between gender and confirmation bias.

Table 4.8

ConJrmnir n bias with rexyect to 8ender

Con6nzetaoa Pearson Chi Asym Sig (2-

Gender
Susceptible (9â) fosusceptible (9F) Square tail‹xi)
Male
29.SP 70.5
F-'emcle

37
Table 4.9 explains the relationship. between gender and status quo bias. About half of the
respondents (57.4&) have shows susceptibility for status quo bias while remaining
proportion of 42.6 percent of the respondent has not shown susceptibility for this particular
bias. The proportion of suscepñbility is also similar in case of male and female investor. fn
both of the cases, about half of the respondent has shown the susceptibility. Likewise, there
exists no significant relationship between gender and status Quo Bias.

Table 4.9

Status quo bias with respect to tender

Stdm o Peason Chi Asym. Sig (2-
Gender Suw epUA e(0b) InsuwgpCble(96j Square tailed).
Male 56.89b 43.2?b
Peoude 58.5 41.SP 0.034 0.855

This section explains the influence of age on different behavioral bias. This demographic
variable was divided in three category of analysis. They are 20 to 30, 30 to 40 and 40 and
above.

Table 4.10 prments the relationship between age and loss aversion bias. The respondents
Of ilgt• C8t8gOJ Of 2[) ID 3(i df8 tllOYR gU8D£• fi EU FOR 1065 dV£'4T› folt bfaS IR COlft df£i 10 OfhBf

age category. Out of 136 respondents, 82.1 percent of the respondents from 20 to 30 age
category, 63.2 percent of respondent from the age category of 30 to 40 and 71.4 percent of
respondent frnm the age category of 4D and above are susceptible for loss aversion bias.
The test of significance showed that there exisn no significant relañonship between age
and loss aversion bias.

38
Table 4.10

Loss aversion bins with respect to age

Los4 Aversion Pearson Cbi Asp
Age Siisoeptible (&)
Square Sig (2-

20-H 17.9&
30-40 368W 4.330
40 and above 286W

Table 4.11 exhibits tlse influence of age on overconfidence bias. The respondent of age
category 20 to 30 and 30 to 40 are more confident in compare to the age category nf 40 and
above. Mnre than 80 percent of the respondents of age category of 20 to 30 have shown
overconfidence bias. which is higher than the age category of 30 to 40 and 40 and above.
There is no significant relationship between age and overconfidence bias.

Table 4.11

Overcoyfidence bias with respect so age

Age Susceptible (&) Insmme bid(W)
OvnoooGd

20-H g2.1 P6 17.9&

30-40 78,99é I .634 0.442
4O und above 71.4 28.6&

Table 4.12 shows the influence of age on optimism level. The table explains that the

eaiegory. The respondent above the age of 40 are least optimist and hence they are least
susceptible to this bias. Less than two third (66.7&) of the respondent of above 40 are
susceptible to this bias.

39
Table 4.12

Optimism bia.s with respect to age

Opbmism Pearson Clu Asp
Age @soeptible (&) Square Sig (2-

20-H 82.1 & 17.9&

30-40 73.79b 2634 t .634 0.442
40 and above 66.7& 333&

Unlike overconfidence and optimism bias, mental accounting bias has been seen more in
the age category of 30 to 40. As per table 4.13, almost 90 percent of the respondent of age
category of 30 to 40 has shown mental accounting bias while 67.9 percent and 71.4 percent
of the respondent has shown the bias in their respective age category of 20 to 30 and 40
and above. Unlike in other test of significance, there exists s ’r t relationship between
age and mentor. accounting bias.

Table 4.13

Mental ac counting bias with respect to age

Susceptible (&) Insinocptibic (&) Sqm Sig (2-
Mental Accounting

67.9&
3 -40 89.S9â USA 6.055 0.048
40 and above 7 l.4& 28.6W

Table 4.14 reports that the respondents frnrn age categnry of 40 and above are less
susceptible to the ilhision of control bias. Our of all respondent of this age category, only
14.3 percent of the respondent has shown this bias. The age category of 20 to 30 is more
susceptible to this bias among the entire three age category. Aiso, the test shows that there
exists no significant relationship beiween age and illusion of cnntrnl.

40
Table 4.14

Illusion of control bias with respect to age

illusion oftootroi Pearson tIu Asp
Age Siisoeptible (&) 1nsusceptibk• f@) Square Sig (2-

20-H 32.1 & 67.9&

30-40 26,3& 73.†& 4.10 0.127
40 and above 14.3& 85.7&

In the test of confirmation bias among the respondents, the respondents nf age category of
30 to 40 were less susceptible to this bias among the three age category. Only 21.1 percent
of the respondent of the age category 30 to 40 has been influenced by confirmation bias,
The age category of 40 and above is more influenced by this bias. There exists no
significant relation between age and confirmation. bian.

Table 4.15

Confirmation bias with respect to age

Age Sueceptible (%) lnemoe ibB( 1

28.6% 7t.4%
3&40 2I.I @ 78.9& 4.682 0.096
40 azid abnve 4296 57.1&

Table 4.16 illustrates the influence of age on status quo bias. The respondents of age
caiegory 20 to 30 are more susceptible for siatus quo bias. The age caiegories of
respondent of 40 and above are As susceptible i.e. 42.9 percent of the respondents are
susceptible for this bias. The test of égnificance was nut similar with other previous test.
In this case there is significant relationship between age and status quo bias.

41
Table 4.16

Status quo bias with respeti to

age
S.tatus Quo Pearson Cbi Asp
Age Siisoeptible (&) lnsusceptibie (&) Square Sig (2-

20-H 67.9'B 32.1&

30-40 57.99b 42.1& 6.139 0.046
40 and above 42.9& 57.1&

4• •3 Behavioral bias with respect to educational background

This section explains ihe influence of educational background on different behavioral bias.
The educational background was categorized as finance and non-finance.

Table 4.17 shows the influence of educañonal background on loss aversion bias. The
collected data shows that the respondents from finance background are more suscepoble
for loss aversion bias. Almost 80 percent of the respondents from finance background are
loss averse white about 65.5 percent of the respondent who are from non finance
background has ahown loss aversion bias. There exists no significant relationship between
educational backgmund and loss aversion bias.

Table 4.17

Mss Aversion bias wit/t resPe ct to educational background

Educational
Lnss Avenion Pearson Chi Asym Fig (2-
Background
Suuepiible t&) lnsincepiible.t&) Squnm tniloJ)
Pinanre
79.5& 20.59£
Non Fimnce
65J 34.SP 3.335

Similarly, Table 4.18 presents the influence of educational background on overconfidence

bias. The respondent of finance background has shown more of overmnfidence bias than
those of non finance background. Out of the respondents from finance background, 89.7
percent of the respondents have shown overconfidence bias while out of non-finance

42
respondents, 62.1 percent of the respondents are in the influence of overeonfidence. The
test showed that there exists significant relationship between educational background and
overconfidence bias.

Table 4.18

Overconfidence bias with respect to educanonul background

Eclucat anal Uverconfidence P on Clu Asym. Sig (2-
Background Susceptible (96) Insusceptiblc (%) Square tailed)
hinztncc 89.7% i0.396
Non Finance 62.IN 37.99a 14.8 18 0.00

The result for inHuence of educational background on optimism bias was similar with
overconfidence bias. In this case also, the finance background respondents are more
optimist than the respondent from non-finance background. The proportions of around 80
percent frnrn finance background are opomiat in compare to 69 percent of respondent from
non finance background. This can be seen in the table 4.19. There exists no significant
relationship between edumtional background and optimism bias,

Table 4.19

OPtimism bias with respect to educational background

Educational Pearson Chi Asym. Sig (2-
Background Susceptible (m) Irixuscepñble (&) Square tailed)
Finance 79.59b 20.5
Non Finance 69.09F 31.0& 1.964 0. 161

Table 4 20 expIain8 that Eoth finance and non finance background respondent are almosi
equally susceptible for mental accounting bias. Around 75 percent of the respondent frnm
both education background caiegory were susceptible for the mental accounting bias. Also,
the table shows that there exists no significant relatinnsbip between educational
background and mental accounting bias.
Table 4.20

Mental éccoiinring bias with respect to educational hack3round

Educational Mental Accounting Pearson Chi Asym Fig (2-
Background Susceptible (&) lnsuscepñble (%) Square tailed)
Pinance 74.4& 25.6
75.W 24.IN 0.40 0.841

Likewise Table 4.21 reports that respondent from non finance background are more
susceptible to illusion of control bias than the respondent from finance background.
Furthermore, there exists no significant relationship between educational background and
illusion of conirol bias.

Table 4.21

illusion of control bias with respect to educational background

Rducationai
Illusion of Control Pmrson Chi Asy m. Sig t2-
Background Susceptible IN) Insusccptiblc UI Square tailedj
Finance 23.1& 76.9&
Non Finance 27.69b 72.4& 0.361 0.348

In case of confirmation bias also, non-finance background respondent are more susceptible to
the bias. This is illustrated in table 4.22. About 28.2 percent of the respondents from finance
backgmund are susceptible to this bias while 34.5 percent of the respondents frnm non
finance background hold confirmation bias. Likewise, there exists no significant relation
fitween educational background and confirmation bias,

Table 4.22

Confirmation bias with reject toedscational ba‹-k8 round

Edue niional
Confimotion Pearson Chi Aaym. Sig
Background
8usceplible (&) lnsuscepuble (2- Square tailed)
(&) Finance
Non Finance
0.614 0,433
Table 4.23 shows the influence of educational background on status quo bias. The
collected data shows that again non-finance respondent are more susceptible to the bias.
There exists no significant relation between educational background and status quo bias.

Table 4.23

Status quo bias with reject to educational background

Educational
Status Quo Pearson Chi A8yro. Sig (2-
Background Susceptible t&) Insusceptible t&) Square tailed)
Pinance 51.39b 48.7&
65.5 34.5& 2.756 0W7

4.EA Behavioral bias with respect to eccupatien

This section explains the impact on behavioral bias on the basis of four occupation
category. The occupation category has been divided as service, business, student and
investment trader. Investment traders are those who are completely engaged in stock
trading business.

Table. 4.24 explains the loss aversion bias on ihe basis of the occupation category. The
table explains that the raost susceptible group for loss aversion is student group. Among
the studBHtB, 80 percent has shown the While the investment traders are less
susceptible from the bias. Among ihem, 66.7 percent are only stiscepiible which is is in
compared to the other occupation category. The table shows that there exists no significant
relationship between loss aversion bias and occupation category.

Table 4.24

Loss aversion bias with respect to occupation

Lose Aversion
Pearson Cki Asym. Sig (2-
Occupstian Suweptible IN) Irisuscepñble IN) Square tailed)
Service 6fl.8 31.296
78.6& 21.4& 2.473 0.480
W.M X.OF
66.7 333W

45
It can be clearly seen in the table 4.25 that the respondents from service category are found
less susceptible from this bias while student seems to be highly influenced by
overconfidence bias. Among the service caiegmy, 68.8 percent are susceptible while 90
percent of the students are susceptibk from this bias. Again the relationship was not
significant between overconfidence and occupation.

Table 4,25

erconfidence bias i+'ith recfiiect to occupation

Overconfidence
Occupation Susm@Me (%) fnumedbe OQ Square lailcd)
Service 68.896 3L2%
7I.4% 28.6% 5.647 0.143
9&fH£ BOOB
1oyestaicnt Trader 7786 22.2%

As per the Table 4.26, Investment traders are less influenced by the optimism bias while
respondent from business category are most influenced by this bias. Out of the inventment
trader, 55.6 percent are only influenced by ihis bias while more than 90 percent of the
business respondena are susceptible from this bias. There exisis no significant relationship
between optimism bias and occupation.

Table 4.26

Optimism bias watt respect to occupat ion

Optimism
Pearson Chi Asym. Sig (2-
Occt5ieHon
Susceptible (&) Insumeptible (&) Square tailed)
73.09t 25.09b
Business
92.9& 7.1& 12.554 0.06
Studeiu
80.09F 20.0&
Investment Tm4er
55.69b 44.496

Table 4.27 explains that more than 90 percent of the. service holder maintained mental
compartment in the process of making investment While the respondents carrying out

46
business and investing in stock market are less susceptible. froth this bias. In this case. also,
diere is no significant relationship between mental Accounting and occupation.

Table 4.27

MenIul accounting bins with respect to occupation

Mental Accounting
Pearson Chi Asym. dig (2-

Occupauolt Susceptible t&) Insuweptible t&) Square tailed)

Service 93.8& 6.2&
Boeincss 64.3& 35.7 9.996 0.19
Srudeni
6fi.09b 3s.OF
Investment Trader 77.8& 22.2&

Table 4.28 illusirates the level on illusion of control bias or the occupation category. The level
of susceptibility for tbis bias is less in all the four category. The leasi susceptible are the
invesiment waders. Only 16.7 percent of ihe investment traders are susceptible frnm this
bias. The result for rest for significance showed no significant relationship bemeen illusion of
control bias and occupation.

Table 4.28

illusion of comtrol bias with recpect ta accupatian

I Hutton of Control
Pearson Chi Asym. Fig (2-
Ooctrpatiorl Susceptible (&) Inxusceptible ( ) Square tailed)
Service 25.09b 7s.DC
Business 35.79F 64.3& 3.0t8 0,438
Stu&ni 25.W 75.0
divestment Troder 16.7& 83.356

The relationship between confirmation bias and occupation is prmented by table 4.29. The
table shows that the influence nf this bias is also low in the entire occupation category.
Investment traders are mosi influenced by this bias. About 40 percent of the investment
trader has shown this bias white the less susceptible were the respondent carrying put the
business. There exists no significant relationship between confirmation bias and
occupauon.

47
Table 4.29

Confirmation hias wit,h respect to ncupai/on

Confumntion
Pearson Chi Asym Fig (2-
Occupation Susceptible (&) lnsusceptible (H) Square tailed)
Service 31.2& 68.8
Bridness 21.48F 78.69b 2.270 0.510
$tuden\ .30.0'lb 70.OF
Investment Tmder 38.9 61.1 &

Table 4.30 reports that those investment traders are less exposed to status quo bits while
other occupation categories are almusi equal} Sfi6CDQ1›bIe from this bias. Nearly 4n
percent of the investment trader has shown slams quo. bias. In this case also, the test of
significance showed rio significant relationship.

Table 4.30

5lnnis quo bi‹i i with respect to occupation

Status Qtm Bias

Pearson Chi Asym. Sig t2-
8uscepiible (&) lnsusceplible (%) Square tailed)
â2.SP 37.5&
Business 64.3& 35.7& 6.810. 0.076
65.OF 35.OF
Investment Trader 38.W 61.1 &

4.5 Behavioral biases with respect to experience and frequency of trading

In this section, influencing level of behavioral bias will be investigated on the basis of risk
profile, trading frequency and experience in siock investment. These information’s have
been collected through questionnaire and against these variable, behavioral bias will be
investigated.

48
4.5.1 Behavioral biases susceptibifity or the basis of frequency of trading

Table 4.31 illusirates the suscepfibiliiy of respondent against the behavioral bias on the
basis of frequency of trading. The frequency of trading of respondent has been categorized
as regular and irregular. Table explains that the behavioral bias as overconfidence,
optimism, loss aversion, illusion of control is seen mnre in the investor who trade in the
regular basis. Likewise, mental accounting, confirmation bias and status quo bias is seen
more in case of irregular trader.

Table 4.31

Behavioral biases susceptibility on the basis of frequency of trading

Behavioral Biases
OVC OPT MA ILC CON SQ
Irregular 71.gI 68.9% 71.1& 75.6& 254& 33.3 66.7&
Regular 78.39a 95.7% 82.6& 73.9W 26.1& 26. I 39.IN

4.5 s Behavioral binses susceptibility on the basis of experience in stock trading

Table 4.32 exhibits the susceptibility of behavioral bias on the basis of experience in siock
trading. The experience has been categorized in four categories.

As the experience increases, the level of bias getn decrease. When the experience is on
lower aide, investors are more loss averse but as the experience increases, the degree of
loss aversion also decreases. Similarly, during the initial phase investor do hale little level
of overconfidence and optimism bias but the degree of bias as experience
increases and the level again decreases as the experience increases. In case of mental
accounting, the level of influence has increased as the age has increased. There is mixed
result in case of confirmation bias and status quo bias. A strong concrete result cannot be
obtained in case of confirmation and status quo bias.

49
Table 4d2

Behavioral biases suareptlblllty on the basia of experience In tredlng

OVC OPT EA PLC CON S.Q

0-2 87. I& 67.7& 74.2& 71.OF 32£96 25.8 54.8&
2—4 71.49b 1tD.0& 8S.†& 57.1& 2fi.69b 42,d 71.4&
4-6 80M 80.0G 60.0'ñ› 0. 20.W 20.OF
6+ 55.W 77.8& 66.7& HD.OF 16.79F 33.3& 61.l&

4.6 Measurement of reliability test using cronbach’s alpha

TaYakol and Dennick (2011) conceptualized ihat alpha is an important concept in the
evaluation of assessments and questionnaires. It is mandatory thai assessors and researchers
should estimate this quantity to add validity and accuracy to the interpretation of their data.
Nevertheless alpha has frequently been reported in an uncritical way and without adequate
understanding and interpretation. High quality ttsts arc important to evaluate the reliability of
data supplied in an examination or a research study. Alpha is a commonly employed index of
test reliability. Alpha is affected by the test length and dimensionality. Alpha as an index of
reliability sbould follow the assumptions of the essentially tau-equivalent approach. A low
alpha appears if these assumptions are not meei. A longer tesi increases the reliability of a
rest regardless of whether the test is homogenous or not. A high value of alpha (> 0.90) may
suggest redundancies and show that the test length should be shortened.

Kline (2000) proposed a commonly accepted rule for describing internal consistency using
Cronbach’s alpha is as follows, however, a greater number of items in the test can artificially
inflate the value of alpha and a sample with a narrow range can deflate it, so this ruie should
be used with caution:

50
Table 4.33

Cronbach alpha value

Cronbacb’s alplm lotexnel conelsteztcy

? 09 Bsceiletit {I-figh-Stakes feeling)
0 7S n < 0.9 Good (Low-Siakes testing)
0.6 a < 0.7 Acceptable
0..5S n < 0.6 Poor
< 0.5 Unwveptable

The reliability test i.e. calculation of alpha value for likert scale question is carried out. The
decision making statements of all the behavioral bias are calculated and presented below.

Table 4.34

Cronbach alfiihu yalue of decisian makiny stc/1exoent

Liket1 Scale SLBt ment Number of Statenu•nt Alpha Vaiue

s aversion bias decision making etatements 3 0.71 S
Owrconfidsnce bias decision making slatemen 3 0.817
OpLinusra bias decision making statements 1 0 717
Mental accounting bias decñion eiakinp statements 2 0.785
Illusion of control bias decision making statements 2 0.667
Confirrrettion bias dnision making statements 2 0,63i

Smtus quo bias decision making statements 3 0.822

Complete Rei uf decision making stntcrmnts J8 0.881

Table 4.34 shows. the alpha value for the test of reliability of decision making statement for
the respective behavioral bias. From the table it can be 6een that beside illusion of control
and Confirmation bias, ali other five behavioral bias have alpha value more than 0.7. Thus
we can conclude that the decision making statement for loss aversion, overconfihence,
optimism, mentai accounting and status quo bias is consistent with the criteria stated by
Kline. Likewise, the alpha value of illusion of control and confirmation bias is between 0.6
and 0.7 u'hich are also acceptable. This concludes that the likeri scale decision making
questionnaires are reliable.

51
Similarly, on n whole the alpha value while taking all decision making statement also falls
in reliable category.

4.7 Influence level of behavioral bias on investment decision for tnvestors

The influence of behavioral bias on financial decision making is determined by the mean
value obtained from the likert scale statements for each behavioral bias separately. The
likert scale statements are anached on the appendix. For likert scale statement, five point
scale has been used. This determines the intensity of influence of behavioral bias on
financial decision making The likert scale statement ranges from strongly disagree to
strongly agree. The factors in the liken scale were rated against these dimensions and
them dimension with their coding are as follows:

• 1- Strongly Disagree
• 2- Disagree
• 3- Neutral
• 4- Agree
• 5- Strongly Agree

In order to test the influence nf behaviors bias on the decision making, two or three
statement in likert scale were used for each behavioral bias. Three decision making
statements were used fur loss Aversion, nverconfiderice, optimism and status quo bias.
Similarly, two decision making statements were used for ilhision of control, confirmation
and mental accounting. Afier the daia was collected, the mean values for these statements
were aggregaied.

The midpoint value the likert scale statement is 3. The value above this midpoint signifies
higher influence while the value below this midpoint value signifies lower inlhience on
decision making. Table shows the aggregated decision making statement descripove
analysis. The mean value and standard deviation value is presented in the table.

52
From the table 4.35 it can be inferred that the mean value for all the behavioral bias except for
status quo bias is above midpoint value 3. The mean value of status quo bias is below
3. This shows that status quo bias has lower influence in the financial decision making
process. Likewise, the other bias i.e loss aversion, overconfidence, optimism, illusion of
control and confirmation bias has higher level of influence in the financial decis ion
making. The mean value for loss aversion bias is 3.0735. From this it can be inferred that
loss aversion bits has moderate level of influence in the financier decision making
process of Nepalese investor.

Table 4.35

Aggregated Statements Mean Std. Deviation

Aggregated losr aversion decision making statement 3.0T35 .9320ñ

A@gepted ovbtodñde e Ueciaoo'" "ss°* "' 3.6324 .9413 I

Aggregated opti mism dwision making statement

Aggregated meñtci accounting decision n iking statement 3.66l I .9523.5

Ag qamdilxionof?ontmldrcionmaugaazmcm 3.6324 .98739

Aggregated confirmation decision making statement 3.7TOQ .964J7

A8gre8atcd status quo decision making statement 2.9412 &7572

Also, it can be concluded that confirmation bias has higher degree of infiuenoe in the
decisinn making since it has the highest mean value. Starus quo bias has lnwest level of
influence among the tesied behavioral bias.

4.8 Relutiooship between behavioral biased and financial decision making

The relationship between the behavioral bias and finariciai decision making can be
established by the study of correlaiion betz'een these two variables. Investment decision is
rakes «s dependent variable while behavioral biases are takerl as independent variable.

53
Since there were. use of two .or .three decision moking statement to test the relationship
between behavioral bias and financial decision making, these statement were aggregated.
AH the behavioral bias are taken as independent variable separately and their consecutive
decision making statement are taken as dependent variable.

Table 4.36 repusents the relationship between behavioral bias and decision making. The

been .done at 5 percent and I percent significance level.

Table 4.36

Relationship between behavioral bias and financial decision mHing

Behavioral Biases Pearson Correlation Fig t2 — tailed)
Loss aversion 0.214 0.017*
Overconfidenoe -0.397 0.O30**
Optimism 0.140 n. lf13
Menml afmunting -0.45 0.605
Illusion of coniroi 0.26 0.765
Confirmation 0.191 0,026*
Status qtm 0.024 0.781
**. Correlation is significant at the 0.01 level (2-failed).
*, Correlation is significani at the 0.05 level t2-uiled).

From the table, it can be seen that there exists positive relationship between loss aversion
and financial decision making and is statistically significant. This implies that investment
decisiona of Nepalese investor are affected by loss aversion behavior and Nepalese
investors basically invest in sure gain than the probable gain. Likewise, for overconFideace
bias and financial decision making the relationship is significant. The overconfidence
behavior of Nepalese behavior can be seen in their decision making. Similarly, the
relationship between coefirmation bia8 and financial decision making is also significant.
Nepalese behavior want to hear positive and favorable news and their decision making is
influenced by this bias.

The relationship between oilier remaining bias and financial decision making are
insignificant. These other variable includes optimism, mental accounting, illusion of
control and status quo bias. Although optimism, mental accounting and status quo bias has

54
been shown by majority of respondents but these biases are not correlated with financial
decision making.

The study of interrelañonship between the selected behavioral bioses is one of the major
objectives of the study, One behavioral bias can influence another bias. The relationship
between the behavioral biases is studied by analyzing the correlation between the
biased. Table 4.37 exhibits the correlation between the biases for the study. The
correlation between the biases has been studied at the significance level of 1 percent arid
5 percent significance level.

Tire data presented in table shows thai there esists ten significant relationships. The first
significant relationship exists between loss aversion and optimism bias. The relationship
is positive and significant. Similarly, loss aversion has significant relationship with
raentai accounfing, confirmañon and status quo bias but the reletionship is negative.

Similarly, overconfidence bias aisn has significant relationship with optimism bias and
confirmation bias. The relationship with optimism bias is significant at 5 percent and the
relationship is positive. On other hand, the relationship with cnnfirmation bias is negative
and significant at both I percent and 5 pemcnt.

Overconfidence bias also bas significant relationship witb opamiszn bias and coxfirmabon
bias. The relaiionship with optimism bias is significant at 5 percent and the relationship is
positive. The overconfident investor will have more confidence in hislher decision and wilt
trade more and be optimist that he will earn geod return or had taken positive decisinns.
On other hand, the relationship with confirmation bias is negative and significant at both 1
percent and 5 percent. Thus, tne overconfident investor will ignore the negative
information and rely on positives one only.

55
Table 4.37

Relationship between behavioral biases

CERN
1 .083 .346 -P69 .038 -.176 -248
.338 .000 .0J2 .656 .040
OVC 1 .184* .020 -.J02 -.259“
.032 .813 .236 .003 2%

OPT 1 -.176“ .176* -.129

.040 .040 .136

MA .176” .018
.040 .832
1 -.165 .292"
.054 .001
CON 1 .126
.144

SQ 1

*•.. C-lation ix significant at the 0 01 level (2-railed).

*. Cnrrélation is significant at the 0.05 level (2-tailed).

Likewise, optimism biases aiso do have significant relationship with mental accounong
and illusion of contrnl bias. Both the rslatinnship is signi£cant at 5 percent significance
leveL Optimism bias and mental accouniing bias do have negative Nation while there
exists positive relatinn with Illusion of control. The optimist investor will create a separate
mental compartment where the inventor will place those investment which the investor
believe will outperform the marker and in another compartment those investment on which
the investor do not have any faith. Also optimist investor will have wore belief in his
ability to control the return for investment.

Mental accounting and illusion nf conirol also do have positive and significant relationship
at 1 percent significance level. And .the last relationship exists between illusion of control
and status quo bias. The relationship is positive between these two biases.

56
The positive relationship signifies that if the influence of one bias increases, ihe influence
of another bias will also increase. For example, if the influence. of overconfidence
increases then there are chances that the individual will demonstrate optimism bias. Similar
relabon exists between loss aversion and optimism, optimism and illusion of control,
mental accounting and illusion of control, and status quo and Dust on of Contml.

Likewise, negative relation signifies the decrease in influence level with the increase in
influence level of another bias. Such relation exists between loss aversion and mental
accounfing, loss aversion and confirmation, loss aversion and status quo, and optimism and
mentai accounting.

4.10 Regression analysis

For the measurement of influence of behavioral bias on investment decision, regression

analysis has been used in the study. Regression analysis has been done by taking bias
individually and by taking all bias in a whole as well. As mentioned in the theoretical
framework, investment decisions are taken +s dependent variable while behavioral bias is
mkonedo dentveiblc.

Table 4.38 shows the regression analysis between the financial decision making and
behavioral biases. It shows the asonship between behavioral bias and financial
decisions. The table only shows the model which had significant rslati pnship. Ail other
models which were nor aigniticant are not presented in the table.

The table shows thai, loss aversion, overconfidence and optimism bias are iigir»ut «t
either I percent or at 5 percent. In Medel 4, where R square is 0.264, independent
variables are able to explain 94.2 percent of the variability in dependent variable.

The F ratio in the table shows that if the overall regression medel is a good r • not for
the data. The Model 3 shows that the relationship between independent and dependent
variable is statfitically significant. This concludes that there is influence of behavioral bias
on investment decisions.

57
 LA ufi

• (T.B97

a.1967 0.O4 0.036 6.072

5S
Unsfandardized indicate how much dependent variable varies with on
independent variable, where all other independent variable is made constant.

Therefore, loss aversion, overconfidence and optimism bias individually have impact on
financial decisions. Likewise behavioral bias also influences the financial decisions.

4.11Behavioral bias in institutional investor

This section presents the behavioral bias shown by the institutional investor. Interviews
were conducted wiih four employee of institutional investor. The responses were
aggregated, summarized and conclusion is drawn. The section has been divided into three
segments: interviewee profile, analysis of interviewee data and summarization of
interview.

d 11.1 Interviewee profile

The table 4.39 presents details of the interviewees. The interviewees were assigned codes
for further ease of use. Code ‘I’ has been used for denoting interviewee. The table
provides the interviewee name, respective code assigned, organization and designation of
those interviewee.

Table 4.39

Ince n'iewee
profile
Interviewee Code Organization @R6ig£8tiOfl

Janardhan Acahrya I-I Nabil Investment Banking ltd RcrcachAvxioe

8ubnsh Poudel HiBS Capilai Ltd. Officer

I-3 Relationship Officcr

5imant Koimln Sikhnr Insinunce. €lffieer

59
4.11.2 Analysis of interview rmponse

The interviewee were asked a series of quesfinns which were listed in the checklist. The
question pattern followed the approach of moving to specific from general. In the initial
phase, general questions were asked and the questions were made specific in order to meet
the objecñves of the study. The specific questions were targeted to derive required respone.
This section will only focus on those specific questions. All the interviewees were asked
similar questions.

In response to the question regarding how the investment decisions are carried out in their
respective organization, all the respondents presented their response. The responses were
similar i.e. the investment decisions are taken by a team which have specialized knowledge
in the analys is of securities. Initially, securities will be selected then after, purchases and
the securities will be managed which can be referred as portfolio manegement.

Aligning with ihe previous question, the respondents were asked regarding the criteria of
Belecting the securibes. The reapondent’a responses were similar in case of criteria. They
pointed out the use of fundamental and technicN analysis. Furrier question was added in
order to know if there are presence of behavioral and psychological factor in the
investment decision and portfolio management process. The respondent replied that since
the securities are analyzed through fundamental and technical analysis, there is no presence
of psychological or behavioral aspect in investment decision.or portfolio management.

I-4, highlighted a point

“ Although during the decision making yroces.s there will be some aspect az past

decision are made by analyñng the .me rit.s of securities through firmer analyst.s,
yre.hence of such factor cannot be seen in case of final decision made.”

In the similar aspect, the respondents were asked if they preferred to hear positive news
and neglect negative news. The respondents replied that there ia nothing like negative news
for the organization till they analyze the impact of news. They will analyze all iype of
information and investigate if the news will increase or decrease the security price or will

60
have some positive or negative impact in near future. This confirms that the confirmation
bias is nor present in case of institutional investors.

The interview session mov«i Caer and the questions were asked targeting to know the
presence of loss aversion in the institutional investors. The respondents were asked how
the buying and selling of securities is done, what events triggers the buying and selling of
securities and if the security which is making profit or making loss is sol d more. The
respondents replied that their organization have fixed initial rule. There will he certain
margin and if the securifies price falls or rise below ihat margin then the trading of
securities is done. Regarding the question if the losing securities or wining securities are
traded more, respondent replied thai any securities falling under the rule will be traded.

I-3 forwarded that:

“The losing securities will be sold once ii louthes the margin the organizniieri had set and
the winning one will be sold in similar way. However, ij‘the analyst finds mhm the securities
price have reached to its peak and cannot increase more then also the security will be
sold.”

This shows that the institutional invastnr do not exhibit loss aversinn bias. The selling and
buying of securities is done completely on the basis of predefined rule.

Likewise, the next questions were targeted to know the presence of overconfidence and
optimism bias in the institutional investor. The respondents were asked regarding how
much confidence they have in their decision. The msporidents replied that since the
decision are done through proper analysis, use of fundamental and technical analysis and
analyzing the merits so they do have confidence in their decision. They also mentioned that
they are optimist in the decision they made. The level and confidence and optimism are
due to their proper analysis. Although they are confident and optimist but optimism bias
and overconfidence bias cannot be drawn from their responses.

The reapondtnts were asked if the organization determines expected level of return in their
investment and how positive they feel regarding achieving that level of return. The
response was similar as of testing overconfidence bias. The use of proper analysis makes

61
them feel positive io achieve the expected return. This also shows that they are optimist
and confident but not influenced by optimism and overconfidence bias. A further question
was added if the expected return falls below actual return do they have plan or will make
plan to achieve the expected level of rt inn.

1-1 mentioned tlsat:

“If the actual return falls below the expected then the organization will plan to meet tkat
level and the y can cantrol their return through specific planning bW the option ta maintain
the expccced level is limited as all the option do not falls under company objective. But trill
the y try to analyze tM best way to maintain the status. ”

This concludes that the institutional investor also do have status quo bias. Alongside, the
status quo bias, they are also influenced by illusion of control bias.

Likewise, the respondents were also asked regarding the diversification of securities. It

securities. Likewise, they aiso invest in fixed securities of relatively secured securities
also. It can be concluded thai Mental accounting is also prevailed in institutional
investors.

4.11 Summarization of interview rnponse

In this part, the response of interview bas been summarized in terms of presence of bias in
institutional investors. The following point will summarize the interview:

• There is no presence or influence of psychological and behavioral aspect in the.

investment Vision and portfolio management.
• Loss aversion has not been exhibited by institutional investor as buying and selling
of securiaes is done under predefined rule.s.
• Institutional investors are confident and optimist in their decision as the decision is
made through technical and fundamental analysis but on the same hand they are
not influenced by excess optimism bias and overconfidence bias.
• Institutional investors try to maintain their status quo and belief that they have
control on their return.

62
 • Mental accounting is also prevailed in institutional investors as they segregate the
investment and also invest in those securities which are relatively safe.

4.12 Findings

The major findings of the study after the analysis of data are as follows:

• The stock market is dominated by mate investors. About 70 percent of the investor
are male while other remaining are female.
• Individual from financial background invests more in the stock market in compared
tn the individual from non finance background.
• There is significant proportion nf individuals who are completely engaged in stuck
investment. This propmtion is almost one fourih of iotal investor.
• Most of the investors are either risk averse or risk neutral in nature.
• There is only one third of investor who trade or invest regularly in stock market.
• Investors are more influenced by loss aversion, overconfidence, optimism and
menial accounting bias in compared to illusion of control, confirmation and siatus
quo bias.
• Male are exhibits more loss aversion, overconfidence, optimism and mental
accounting bias. While female ars more influenced by confirmation and status quo

• There was no significant relaiion between gender any of tlse tested bits.
• The age group of 20 to 30 was more influenced by loss aversion, overconfidence,
optimislB and 6tatus quo bias. While the mid age group i.e 30 to 40 was more
influenced by mental accounting bias. Similarly, confirmation bias mosily
influenced the investor above the age group of 40.
• There was significant relationship between age and mental accounting and status
quo bias.
• Beside Loss aversion, overconfidence and optimism bias, investor from non
finance background were more susceptible to other remaining bias. The
 respondents from finance background were more susceptible to loss aversion,
overconfidence and optimism bias.
• Overconfidence had significant relationship with educational background.
• There was significant relationship between behavioral bias and occupation.
• Regular traders are more loss averse, overconfident, opñmist and have higher
degree of illusion of control. While those who trade in irregular basis are inBuenced
my mental accounting, confirmanon and status quo bias.
• As the investor gets more experienced, the lesser they get influenced by the
behavioral bias. However, in case of mental accounting and status quo bias, same
cannot be said.
• Confirmation bias has higher level of influence in the decision making since it has
the highest mean vaiue. Status quo bias has lowest level of influence among the
tested behavioral bias.
• Loss aversion bias bas moderate level of influence in the financial decision making
process of Nepalese investor.
• Through correlation analysis, it was concluded. thai the. relationship between loss
aversion, overconfi dence and confirmation bias was significant with investment
dcci Biofl. Likewise positive correlation existed for all the bias except
overconfidence and mental accounting bias with investment decisions.
• Correlation was significant between different bias. The significant correlation can
be seen in Table 4.37
• Behavioral biases have significant relationship with the investment decisions.
The cult was obtained through regression analysis
• Loss aversion, overconfidence and optimism bias inaviduaily have significant
relationship with the decision making of investors.
• Mental accounting, illusion of contml and status quo bias are exhibited by the
institutional invesmrs.
 Chapter V

Discussion, Conclusion and Implications

This section presents. the discussion, conclusion and implication that could be drawn from
the study. The section is divided into two segments. In the first segment, the discussion is
carried ouL Likewise, the conclusion and implication is drawn in second.

5.1 Discussions

The finding of the study shows that behavioral bias affects ihe decision making process of
the investors. This outcome has been consistent with the study of Bashir, Saved, Ali,.Meer
and Naseem (2013) and Adhikari (2010). Their study also showed that behavioral bias has
influence on the financial decision making of the investors.

The result from the study showed that the more than the majority of Nepalese investors
were influenced by fuss aversion, overcoiifidence, optimism, mental accounting and ntatus
quo bias. While the illusion of control and confirmation bias affected a smart proportion of
the respnndents, Hnwever it was loss aversion, nverconfidence and nptimism bias which
had significant relationship with the financial decisions. These results were aligned with
the previous studies of Odean (1998) for loss aversion, Chen, Kim, Nofsinger and Rui
(2€El7) for overcorifidence and Kafayat (2014) for optimism bias. Also the study of
Barandngh and Hasonzadeh (2013) was aligned with current study as both study showed
no influence of mental accounting on investment decisions.

The current study also concluded that male investors are more loss averse: This finding
was not consistent with the previous study of Onsomu (2D14) and Hnsan, Khalid and
Habin (2014). Similarly, Onsomu tesied overconfidence and ihe result was similar with the
current study. In both the study male were more overconfident than the females. The result
for confirmation bias in gender was aiso consistent with Onsomu study, where in both case
female are more affected by Confirmation bias.

Likewise, the investors from finance backgcovod are more prone to overconfidence and
optimism bias. The likely reason may be the capability of the investor which they obtained

65
 from their educational background. The investors whn trade regularly are influenced by
overconfidence and optimism bias. The similar result was produced by the study of Barber
that overconfident investor trade too much and are
optimist in their decision.

The interview carried out with the institutional investor showed that institutional investor
possessed mentai accounting, illusinn of control and status quo bias. The ourput was again
similar with of Fisher arid Statman (1997) who reported that institutional investor are

institutional investor, Mentai accounting and Status quo bias were common. Significant
proportions (more than SOA) of individual investors were influenced by ihese biases.

All the objectives. were fulfilled of the study. The biases were held by both individual and
institutional investor were anai n a• significant relationship between behavioral bias
and investment decisions were identirn, < relationship nf bias with respect to
demographic variables were also studies and the comparison of bias held by individual and
institutional investor was also carried out.

5.2 Conclusions and Implications

The demograpbic profile of Nepalese investor remains same as it was on the pasi. The
market is dominated by the male. The similar respondent prpfile was seen in the study of
Adhikari (2010). The profile in terms of age group was also the same. Likewise, in this
study education background was classified as finance and non finance which cannot be
seen in previous study. It is interesting to analyze the result that the respondenu from non
finance were more susceptible to the bias. However, the respondents from finance
background were influenced by loss aversion, overconfidence and optimism bias. The
reason may be the knowledge of market and finance might had increased the level of
confidence and optimism.

The findings show that Nepalese investor is more prone to loss aversion, overconfidence,
nptimisui, status quo and mental accnunting bias. This concludes that Nepalese investor

S6
prefer to hold the loosing investment while they trade ihe winning one. This goes against
the risk-return principle of finance. Similarly, the reault has shown that Nepalese are
overcorifident and optimist but they have no control on their decLsions as they are not
susceptible of illusion of control bias. This shoW6. thdt Nepalese believe in other random
chances for the return rather than their skills. Also, Nepalese investors divide their
investment into different categories. This tendency is shown by the institutional investors
also.

Although, Nepalese investor has shown these biases but the biases which affects their
decision making are loss aversion, overconfidence and optimism bias. Status quo bias does
not influence the invesonent decisions of the Nepalese investor. Among Loss aversion,
overconfidencn and optimism bias, overcorifidence have higher level of influence followed
by the optimism bias.

Male exhibits more loss aversion, overconfidence, optimism and mental accounting bias.
While female are more influenced by confirmation and stains quo bias. Likewise as the
experience in trading increases the investor are less prone to the bias. The susceptibility for
the bias has decreased as the experience of investor has increased.

As a whole, it can t›e concluded that Nepalese investor are prone to number of biases bui
all the bias do not infiuenne the decisinn making process of the investor.

There are number of implications that can be dmwn from the study, lnvtston have to
know the paCt that market is groom by psychology. There wilt be psychological factor in
ifie market so proper analysis of fundamental and technical tool is important alongside the
understanding of psychological facior. The investor has to be aware about the behavioral
bias inherent in him to tise possible extent. Also the result has shown that the investors
from non finance background are more influenced by the bias so it is very important for the
investor to have financial knowledge while investing in the financial market. Also, the
individual investor can follow the institutional investor as it seen that they are less
inHuenced by the biases and hence their decision can be wiser in compared to other. The
other implications which can be drawn is for the young trader. They are more influenced

S7
by bias so they have to gain more experience in order to eliminate behavioral bias to the
possible extent.

The current study has left door open for further study. Pompian (2£D6) has identified 20
behavioral biases. Theae biases can be rested under the similar study. Also, various other
studies have included the aspect of personality which affects the behavioral bias and
decision making. This also can be study under the similar study with new model. The are
other similar study aiso where the impact of behavioral bias on security prices bas been
tested. Such study can also be carried out. Also, the secondary data from the stock market
can be used for further study. Lastly, the study can be carried out in other market as bond
market, real estate, quid or foreign currency market,