Professional Documents
Culture Documents
Determinants of Individual Investor Behaviour: An Orthogonal Linear Transformation Approach
Determinants of Individual Investor Behaviour: An Orthogonal Linear Transformation Approach
13 January 2011
Online at https://mpra.ub.uni-muenchen.de/29722/
MPRA Paper No. 29722, posted 22 Mar 2011 16:11 UTC
Determinants of Individual Investor Behaviour: An Orthogonal Linear
Transformation Approach
Abstract
E xpected utility theory views the individual investment decision as a tradeoff between immediate
consumption and deferred consumption. But individuals do not always prefer according to the
classical theory of economics. Recent studies on individual investor behavior have shown that they
do not act in a rational manner, rather several factors influences their investment decisions in stock
market. The present study considers this theory of irrationality of individual investors and
investigates into their behaviour relating to investment decisions. We examine whether some
psychological and contextual factors affect individual investor behaviour and if yes which factors
influences most. E xtrapolating from previous literature on economics, finance and psychology, we
surveyed individual investors to find what and to what extent affects their investment behaviour.
Our conceptual analysis, empirical findings and the perspective framework that we have developed
in the present study, provide five major factors that can influence individual investor behaviour in
Indian stock market. The findings can be useful in profiling individual investors and designing
appropriate investment strategies according to their personal characteristics, thereby enabling them
optimum return on their investments.
Keywords: Individual investor, Psychological biases, Investment behaviour, Indian stock market,
Behavioural economics.
The present paper is partly based on the research work carried out by Abhijeet Chandra for his doctoral study at the
Department of Commerce & Business Studies, Jamia Millia Islamia, New Delhi-110 025, India. Authors would like to
thank all the survey participants for their support and also to the Faculty of Social Sciences, Jamia Millia Islamia, New
Delhi for financial assistance. The views expressed in the present paper are solely those of the authors and should not be
taken as representative of the institutions they belong to.
[1]
INTRODUCTION
For a long time, the investors’ full rationality was the main hypothesis of the most academic research
in finance. In fact, it was mainly supposed that stock prices are fixed by rational investors’
anticipations and reactions. Rationality here refers to the two main factors, namely, the exhaustive
and objective treatment of available as well as potential information. Because of its simplicity and its
success to capture the stock price movements, this famous investor’s rationality hypothesis was for a
long time supported by the academic researchers in finance. Nevertheless and since recent
movements, the financial academic researchers’ enthusiasm for this hypothesis becomes much
weaker. This changing perceptions lead to experimental research by the psychologists by introducing
the irrationality of human beings (von Neumann and Morgenstern, 1944). Researchers in finance
were then motivated to break with the full rationality hypothesis and to recognise from now on the
neutral effect of some psychological biases on the investors’ decisions and reactions, and
subsequently the effect of such reactions on the stock price movements. In the present study, the
researcher makes an attempt in the behavioural finance research. The present study focuses on the
psychological biases influencing individual investors trading in an Indian context.
Recent studies have argued that Prospect Theory (Kahenman and Tversky, 1979) and mental
accounting (Thaler, 1985), apart from several other psychological biases, provide possible
explanations for investor behaviour (e.g. the disposition effect) and for outstanding asset pricing
anomalies such as the equity premium puzzle, the value premium, and the momentum effect. It is
noted that the academics related to this new financial psychological research area are contended with
borrowing and extrapolating the psychological biases from the famous psychologists’ experiments’
results to investors in stock markets. The extrapolation of the psychological results to financial
markets asserts that under the effect of one or a set of natural and human biases, agents in financial
markets could not be of full rationality, especially to understand and react to news immediately and
appropriately. That’s why the agents appear to over or under react to news, driving by this way, a
momentum effect in stock prices, and subsequently in returns.
A critical analysis of the literature prevailing in the subject clearly depicts that individual investor
behaviour acts as an important determinant of movements in stock prices and subsequent returns.
[2]
This may also help in revising the asset pricing theories by incorporating behavioural factors into the
existing theories (behavioural asset pricing theories also referred to as Behavioural Asset Pricing
Models are being worked upon for their validity in financial economics literature across the world).
Information has been one of the most important components in determining the behavior of
individuals. In case of their behavior in stock market, it becomes even more critical to access and
incorporate into their decision making updated information included in financial reports, periodical
press releases, media coverage and so on. Researchers have acknowledged the significance of
information factor in individual investment behaviour. E pstein (1994) examined the demand for
social information by individual investors. The results indicate the usefulness of annual reports to
corporate shareholders. The results also indicate a strong demand for information about the product
safety and quality, and about the company’s environmental activities. Furthermore, a majority of the
shareholders surveyed also want the company to report on corporate ethics, employees relations and
community involvement.
Krishnan and Brooker (2002) analyzed the factors influencing the decisions of investor who use
analysts’ recommendations to arrive at a short-term decision to hold or sell a stock. The results
indicate that a strong form of the analyst summary recommendation report, i.e. one with additional
information supporting the analysts’ position further, reduces the disposition error for gains and also
reduces the disposition error for losses as well. Merikas, et. al. (2003) adopted a modified
questionnaire to analyze factors influencing Greek investor behaviour on the Athens Stock
E xchange. The results indicated that individuals base their stock purchase decisions on economic
criteria combined with other diverse variables. The authors did not rely on a single integrated
approach, but rather on many categories of factors. The results also revealed that there is a certain
degree of correlation between the factors that behavioural finance theory and previous empirical
evidence identify as the influencing factors for the average equity investor, and the individual
[3]
behaviour of active investors in the Athens Stock E xchange (ASE ) influencing by the overall trends
prevailing at the time of the survey in the ASE .
Fisher and Statman (2000) revealed that the sentiment of Wall Street strategists is unrelated to the
sentiment of individual investors or that of newsletter writer (another category of investors provided
by them), although the sentiment of the individual investors and newsletter writers groups is closely
related. They concluded that sentiment can be useful for tactical asset allocation, and that a negative
relationship between the sentiment of each of these three groups and future stock returns, and the
relationship is strategically significant for Wall Street strategists and individual investors. Malmendier
and Shanthikumar (2003) tried to answer the question: Are small investors naïve? They found that
large investors generate abnormal volumes of buyer initiated trades after a positive recommendation
only if the analyst is unaffiliated. Small traders exert abnormal buy pressure after all positive
recommendations, including those of affiliated analysts. Using the NYSE Traders and Quotations
Database, they found that large traders adjust their trading response downward. Hodge (2003)
analyzed investors’ perceptions of earnings quality, auditor independence, and the usefulness of
audited financial information. He concluded that lower perceptions of earnings quality are associated
with greater reliance on a firm’s audited financial statements and fundamental analysis of those
statements when making investment decisions.
Individual behavior can lead them act in a totally unexpected way and subsequently they end up
seeing their investment performing very poorly in usual case. Kim and Nofsinger (2003) studied
individual investors in the Japanese markets and examine their behaviour and performance. They
used the market level data and found that Japanese investors own risky and high book-to-market
stocks, trade frequently, make poor trading decisions, and buy recent winners. Further, these
behaviours and characteristics appear to vary depending on the bull or bear market conditions. They
observe that it is primarily during a bull market where individuals tend to hold high book-to-market
stocks, as opposed to a bear market where they exhibit an inclination towards high beta stocks.
Overall the poor performance by individual investors can largely be explained by this tendency to
hold value stocks during advancing markets and high risk stocks during declining market. They
conclude that these behaviours reveal at the market level also represents important findings and
hence, become one of the important bases of our study of individual investors in India.
Kadiyala and Rau (2004) investigated investor reaction to corporate event announcements. They
concluded that investors appear to under-react to prior information as well as to information
conveyed by the event, leading to different patterns: return continuations and return reveals, both
documented in long horizon return. They found no support for the overreaction hypothesis. Lim
(2004) tried to test the trading decisions of investors. Using trading records of individual investors,
the study tested whether investors’ trading decisions are influenced by their preferences for framing
gains and losses. The study finds that investors are more likely to bundle sales of losers on the same
day than sale of winners. This result is consistent with the hedonic editing hypothesis, according to
which individuals prefer integrating losses and segregating gains. Alternative explanations based on
[4]
tax-loss selling, margin calls, the number of stocks in the portfolio, the difference in the potential
proceeds from selling winners and losers, correlations among winners and among losers, and delays
in sales order execution do not fully account for the observed behaviour. In addition, the extent to
which mixed sales of winners and losers are consistent with the hedonic editing hypothesis is greater
than what we could expect under random realizations. The evidence suggests that a psychological
error called mental accounting is likely to play a significant role in investors’ trading decisions.
The study conducted by Brown and Cliff (2004) extends much scope for the present study
conducted in the Indian context. In their study, Brown and Cliff (2004) investigated investor
sentiment and its relation to near-term stock market returns. They find that many commonly cited
indirect measures of sentiments are related to direct measures (surveys) of investor sentiment.
However, past market returns are also an important determinant of sentiment. Although sentiment
levels and changes are strongly correlated with contemporaneous market returns, the tests in this
study show that sentiment has little predictive power for near-term future stock returns. Finally, the
evidence does not support the conventional wisdom that sentiment primarily affects individual
investors and small stocks.
Fischer and Gerhardt (2007) conducted extensive research on individual investor investment
decision making. They find that individual investor investment decisions deviate from
recommendations of financial theory. They show that these deviations lead to considerable welfare
losses. Therefore they conclude that financial advice is potentially correcting factor in investment
decision making process and construct a simple model to capture its very impact on individual
investors’ investment success, measured in risk-adjusted return and wealth.
After analysis of the existing literature on individual investor behaviour and economic and
investment decision making, the main issues reported in the above studies can be summarized as
follows:
There is no support for the overreaction hypothesis.
Investor over-reaction to a long series of bad news could produce predictable mispricing of
stock.
Classical wealth-maximization criteria are important to investors.
The recommendations of brokerage houses, individual stock brokers, family members and
co-workers go largely unheeded.
Investors exhibit a strong demand for information about product safety and quality, and
about the company’s environmental activities.
There exist strong forms of the analyst summary recommendation report, i.e. one with
additional information supporting the analysts’ position further, reduces the disposition error
for gains and also reduces the disposition error for losses.
Individual investors are influenced by a number of psychological factors while making
investment decisions.
[5]
The behaviour of individual investors caused by underlying sentiments has a significant
relationship with the movements in stock prices and hence, with the stock returns.
Net trading by individual investors is a powerful predictor of future prices and returns that is
not subsumed by either past returns or past volume.
Most of the studies are carried out in developed economies context. Little evidence has been
available from emerging economies. This aspect particularly makes the present research
study more relevant in Indian context. Studying Indian individual investors trading
behaviour would seem an interesting proposition for both the market stakeholders and the
regulators and policy makers.
All these issues are relevant enough to initiate an empirical study on individual investor behaviour in
an entirely Indian context. Considering these motivations, the researchers attempt to provide
evidence for irrational financial behavior of individual investors.
Another research concern is the role and importance of individual investors and their trading
behaviour in Indian stock market. Unlike institutional investors, individual investors are believed to
be less informed, have psychological biases and also thought of as the proverbial noise traders in the
stock market. The researchers in finance tend to give more importance to the behaviour of
institutional investors rather than that of their individual counterparts as far as their respective role
in affecting stock prices is concerned. It is believed that trading behaviour of individual investors
rarely influences the stock prices. With this perception about the individual investors, majority of
trading strategies and stock market policies are designed and focused to their institutional
counterparts. The attention of policy makers, investment advisors, and related service providers is
attracted towards the institutional investors, thereby ignoring the individual investors’ interests to
some extent.
[6]
Despite the growing interests of finance researchers in this upcoming and relatively new stream
popularly known as behavioural finance, very few studies in India have been undertaken with
reference to the behaviour of individual investors. Individual investors are said to be influenced by
some psychological biases. These biases tend to affect their behaviour in financial decision making
and subsequently their trading behaviour in stock market. It is, therefore, important to identify the
factors most influential to individual trading behaviour. Identifying psychological factors affecting
individual trading behaviour and then confirming the presence of these factors among Indian
individual investors will help establish the fact that Indian individual investors tend to make trading
decisions under the influence of specific psychological biases, and that their trading behaviour is
further having a significant relationship with stock price movements.
The preceding discussions coupled with an extensive literature review help the researcher identify
the following research gaps:
(i) Identifying and confirmation of psychological biases prevailing among Indian investors
is required to establish their respective role in trading behaviour of individual investors in
Indian stock market;
(ii) E xamination of the extent to which Indian individual investors tend to be influenced by
various psychological biases is needed;
(iii) Understanding the individual investor behaviour may further lead to understand the
market microstructure better and shift the focus from institution-centric approach to a
balanced approach (where individual investors are also viewed as a significant player in
stock market).
The framework of this study requires clear definitions in critical realist terms of the objects of
analysis. Here we discuss the formal research objectives in order to fill the gaps identified from the
detailed literature review. The present study bases its analysis with following objectives:
(i) to identify the psychological biases which determine individual investor trading
behaviour;
(ii) to confirm the presence of these psychological biases among Indian individual investors;
(These two objectives mak e this study both of exploratory and confirmatory nature.)
(iii) to examine the extent to which these psychological biases are responsible for individual
investment behaviour.
One of the above-mentioned objectives of this study is to explore the factors influencing the Indian
individual investor behaviour. E xamination of factors influencing the individual investor behaviour
is important for all the stakeholders of the stock market, as an understanding of what affects
investor behaviour and how investors respond to market movements would affect their future plans
and help them/their financial advisors devise appropriate asset allocation strategies to reap the
benefits of equity investments. E ven for companies, identifying the most influencing factors on their
investors’ behaviour would affect their future policies and strategies. Finally for the government,
understanding of such factors influencing the investors’ behaviour would affect the required
[7]
legislation and the additional procedure needed to make the stock market more investor-friendly and
efficient.
The present study examines the factors that appear to exercise the greatest influence on the
individual stock (equity) investors, and includes not only the factors investigated by previous studies
and derived from prevailing behavioural finance theories and contextual psychological experiments,
but also introduces additional factors generated through personal interviews that are supposed to
influence the stockholders’ investment decisions in Indian stock market. Personal interviews were
conducted with appropriate inputs from stock brokers, financial advisors, investment consultants,
and high net-worth individual investors in order to determine the more relevant factors influencing
investor behaviour.
The investigation of academic sources for exploring the behavioural factors influencing investor
behaviour includes the academic research in areas of financial economics, behavioural finance and
psychology. Some of the important works in this regard are done by Kehneman and Tversky (1974),
Denial, Hirshleifer and Subrahmanyam (1998), and Barberis, Shleifer and Vishny (1998). E ach of
these works develops some specific sort of theoretical model that attempts to identify some
psychological biases influencing investor behaviour; these psychological factors had been
extrapolated from the experimental psychological studies. The present study considers eight
behavioural factors suggested by academic sources. These behavioural factors are Representativeness
( Kehneman and Tversky, 1974; Debondt and Thaller, 1985; Barberis, 2001), Overconfidence
(Fischhoff, Solvic and Lichtenstein, 1977; Alpert and Raiffa, 1984; Tversky, 1990; Wood, 1996;
Denial, Hirshleifer and Subrahmanyam, 1998; Gervais et al., 2001a, 2001b; Allen and E vans, 2005),
Anchoring (Shiller, 1998; Lebaron, 1999; E vans, 2002), Gambler’s Fallacy (Kehneman and Tversky,
1974; Gevaris et al., 2001a), Availability bias (E llsberg, 1961; Kehneman and Tversky, 19731; Thaller,
1994; Barberis, 2001), Loss Aversion (Rothschild and Stiglitz, 1970; Kehneman and Tversky, 1979;
Tversky, 1990; Lebaron, 1999; Filbesk et al., 2005), Regret aversion (Shiller, 1998; Lebaron, 1999;
Odean, 1999; Statman, 1999; Fogel and Berry, 2006), and Mental Accounting (Thaler and Shefrin,
1981; Shiller, 1998, Tevrsky, 1999; Rockenbach, 2004).
In addition to these eight behavioural factors extrapolated from academic sources, some other
factors are also considered in this study; these factors are suggested by the professional and
contextual sources which include stock brokers, financial consultants, and investment advisors.
These contextual factors that seem to be influencing individual equity investor behaviour are: (a)
Market share and reputation of the firm, (b) Accounting and financial information, (c) Publicly
available information through various media, (d) Advocate recommendation including that of
brokers, family and friends, and (e) Personal financial need.
The definitive list of factors considered for the present study includes thirteen behavioural-cum-
contextual factors, of which eight psychological biases are suggested by the theory, and five are
contextual to Indian individual equity investors.
[8]
(i) Survey Instrument – In the questionnaire are included simple and direct questions in order to avoid
any confusion on the part of the respondents; each question is based on some specific scenario
relating to stock market investing and equity investment decision-making. These scenarios are very
much similar to the situations faced by investors while investment decision-making in stock market.
Such scenario-based questions help respondents relate themselves to hypothetical situations in stock
market and thus, it would be easier to mark their responses. Since, respondents’ orientation may be
reflected in their answers, they are asked direct questions covering all the shortlisted behavioural
factors. The survey questionnaire consists of four parts, one each for personal information, use of
heuristics, use of prospect theory, and impact of other factors on investment decision-making.
Personal information segment asks the respondents to give details about their name, age, gender,
education, annual income, and annual investment. This information will help the researcher draw
some conclusion on the basis of demographic profile of investor respondents. Second segment, i.e.,
the use of heuristics includes questions covering representativeness, overconfidence, anchoring,
gambler’s fallacy, and availability biases. While third segment on the use of prospect theory covers
loss aversion, regret aversion, and mental accounting biases. Fourth and final segment is based on
other contextual factors suggested by professional and contextual sources, and asks the respondents
to rate the impact of five such factors on their investment decision-making. These five factors are as
discussed above, namely, market share and reputation of the firm, accounting and financial
information, publicly available information through various media, advocate recommendation
including that of brokers, family and friends, and personal financial needs.
In the survey instrument, personal information segment needs personal and demographic details of
the respondents; but next two segments on behavioural factors consist of questions with five-point
Likert scale. The study adopted the five-point Likert scale which seems appropriate and ideal for the
survey instrument used in the present study. In fact, the sample respondents were given scenario-
based questions to which they were asked to mark their response in a range from 1 (least likely) to 5
(most likely). The survey questionnaire was finalised after consulting five experts, three from
academics and two from industry professionals. The validity and reliability of the survey
questionnaire was tested using appropriate statistical methods (Cronbach’s Alpha in this case).
(ii) Target Population – It is useful to remind that the data the researcher was interested with are
the attitude and behaviour of individual equity investors that might be influenced by psychological
biases. These data cannot come from the prices of the stocks, but should come from investors.
Since the present study aims to examine the most influential factors for individual equity investors in
Indian context, the target population for the survey would be obviously the Indian individual equity
investors.
[9]
(iii) Sample and Data – Data were collected via a survey of about 500 individual investors based in
Delhi-NCR. The sample was drawn from the clientele of one of the leading stock brokerage houses†
which provided the researcher with access the contact details of their client base. The researcher
contacted their client and requested them to participate in the study. Out of total, above 350
individual investors agreed to participate in the survey. The questionnaires were then administered to
those individual investors through various media as convenient to them. The questionnaires were
distributed through personal contacts, surface mails, e-mails, and also through the executives of the
participating brokerage house in some cases.
It is hereby important to mention that the present study adopted the convenient sampling technique in
order to collect the survey responses. But the sample was randomly drawn for the purpose of data
collection. The choice of this approach can be explained by following three reasons. First, there is a
large number of individual investors scattered across the coverage area of the survey carried out by
the researcher. It is very difficult to get the exact number of individual investors. Again these
investors are using the stock brokerage and investment advisory services from various stock broking
agents and other similar firms. Many of the investors may be the client of multiple brokerage houses
simultaneously. The researcher, therefore, opted to consult with only one stock brokerage house
with significant client base and good market reputation. Second, the contact details of about 500
individual investors were drawn randomly from the pool of its client base. The only criteria laid
down for an individual investor to be included in the sample was that he/she must have an annual
income of at least Rs. 500,000 at that point of time with a minimum of Rs. 50,000 of investments in
equity. The upper cap of the income as well as equity investment was not set by the researcher.
Finally, many of the investors while contacted at initial phase with a request to participate in the
survey, declined to respond the survey questions. The reluctance by investors could be attributed to
the observation that individual investors tend to be sceptical and perceive the financial matters very
sensitive; they were reluctant to reveal their investments and other financial matters for the sake of
their financial safety. All these factors made the researcher opt for the convenient sampling technique, but
all efforts were taken to keep the sample as random as possible by avoiding any sort of biases
associated with the data collection task.
Reliability of Survey Instruments: The content validity of the survey instrument i.e. the
questionnaire was verified by discussions with five experts, three academicians and two industry
professionals, as it is suggested by Devellis (1991). Accordingly the researcher made changes in
terms of eliminating, adding, or rewording some of the items included in the questionnaire. The
criterion validity and construct validity were tested using correlation analysis. It is widely accepted
that to do a orthogonal rotational transformation analysis variables should correlate fairly well, but
not perfectly. The table of correlations matrix was scanned to check the pattern of relationships. An
examination of the results of correlations suggests that correlations among variables are not
statistically significant enough to indicate any problem among variables themselves. In order to test
† The name of the brokerage house which provided with the details of the individual investors for collecting survey
responses is kept confidential on their request. They do not want to be named as this may affect their market reputation.
[10]
whether there exists any issue relating to correlations among variable, the study used the ‘rule of
thumb’ test suggested by Anderson et al. (1990). They suggest that any correlation coefficient that
exceeds the value of 0.8 indicates a potential problem among variables. The result of correlations
does not show any such problem.
The reliability of the survey instrument was tested with the help of Cronbach’s Alpha method. The
Cronbach’s Alpha method allows us to measure the reliability of different categories. It consists of
estimates of how much variation in scores of different variables is attributable to chance or random
errors (Selltzm, et al., 1976). The Cronbach’s alpha is a measure of internal consistency, that is, how
closely related a set of items are as a group. Cronbach’s alpha can be written as a function of the
number of test items and the average inter-correlation among the items:
. ̅
=
. ( − 1) . ̅
where N is equal to the number of items, i̅ s the average inter-item covariance among the items and
e̅ quals the average variance.
It can be seen from the above function that if number of items are increased, the Cronbach’s alpha
is also increased. Additionally, if the average inter-item correlation is low, alpha will be low. As the
average inter-item correlation increases, Cronbach’s alpha increases as well (holding the number of
items constant). As a general rule, a coefficient greater than or equal to 0.5 is considered acceptable
and a good indication of construct reliability (Nunnally, 1976). The overall Cronbach’s Alpha for the
eight categories is 0.902. The Cronbach’s Alpha shows that these categories for survey instruments
are valid and reliable. This suggests that scales used in survey instruments are unidimensional.
Apart from univariate approach, the researchers applied multivariate techniques to analyze the
survey data. The multivariate analysis consists to combine all the psychological attitudes considered
in the survey questionnaire in order to resume them and to reduce them into few main behavioural
axes that should describe and lead the Indian individual equity investor behaviour. E ach axe would
regroup all the correlated behavioural biases. Such as the main axes resorted would be independent
and sufficient to explain the biggest portion of data variance. To achieve the multivariate analysis,
the study used the Principle Component A nalysis method. The PCA method allows to search for
underlying dimensions in the various sets of variables considered in the questionnaire. The Principle
Component Analysis, popularly known as PCA method, lets us determine which factors or
underlying variables have the greatest impact on the subjects. Other statistical tools are also available
for performing similar analytical functions such as Structural E quation Modelling (SE M), Cluster
Analysis, to group subjects on the basis of various factors; Discriminate Analysis, to establish the
extent of impact of various underlying factors among many others. But the choice of the PCA
method was based on its most suitability for such studies, as suggested by Capon et al. (1994) and
Zoghalami and Matoussi (2009). The appropriateness of this approach was also identified by the
[11]
researcher in one of his earlier studies on the selection behaviour of mutual fund individual investors
(Kumar and Chandra, 2009).
The Principal Component Analysis: The purpose of using principal component analysis is to
identify the most meaningful bais to re-express a data set, i.e. the survey response from sample
individual investors. We expect that this approach will filter out the noise and reveal hidden
structure of individual behaviour.
Let X be the original data set, where each column is a single sample of our data set (i.e. ~ X). In the
present study, X is an m × n matrix where m = 13 and n = 355. Let Y be another m × n matrix
related by a linear transformation P. X is the original data set and Y is a new representation of that
data set.
PX = Y …..(1)
Also let us define the following quantities:
P i are the rows of P
x i are the columns of X (or individual ).
Y i are the columns of Y.
E quation 1 represents a change of basis and thus can have many interpretations.
1. P is a matrix that transforms X into Y .
2. Geometrically, P is a rotation and a stretch which again transforms X into Y .
3. The rows of P , { p1, p2, …, pm} , are set of new basis vectors for expressing the columns of
X.
The latter interpretation is not obvious but can be seen by writing out the explicit dot products of
PX .
⋮
= ⋮ [ ⋯⋯ ]
. ⋯⋯ .
⋮ ⋱ ⋮
= ⋮ ⋱ ⋮
. ⋯⋯ .
The form of each column of Y can be noted as:
.
⋮
= ⋮
.
We recognize that each coefficient of yi is a dot-product of xi with the corresponding value row in
P . in other words, the jth coefficient of coefficient of yi is a projection on to the jth row of P . this is in
fact the very form of an equation where yi is a projection on to the basis of { p1, p2, …, pm} .
Therefore, the rows of P are a new set of basis vectors for representing of columns of X .
[12]
The individual investor behaviour is influenced by host of factors, but considering all those factors
for the purpose of the present study is well beyond the scope and affordability of the researcher.
Here, for the purpose of the present study, 13 commonly used factors were taken into account. The
factors were presented in the form of 17 scenario-based questions with a five-point Likert scale to
collect opinion from sample investors. The principal components analysis was used to reduce the
data collected on 17 variables into smaller number of manageable variables by exploring common
dimensions existing among the variables.
Sample Characteristics: The primary data for the present study consist of the responses from
individual investors collected with the help of structured questionnaires. Since this exercise is what
the success of the present study is greatly dependent on, the emphasis is on collecting data from
diversified group of respondent individual investors. The sample was drawn from the clientele of
one of the leading stock brokerage houses which provided the researcher with access to the contact
details of their client base. The researcher contacted their clients and requested them to participate
in the study. Out of total, more than 350 individual investors agreed to participate in the survey. The
questionnaires were then administered to these individual investors through various media as
convenient to them. The questionnaires were distributed through personal contacts, mails, e-mails,
and also through the executives of the participating brokerage house in some cases.
Survey questionnaire were distributed among about 375 sample individual investors based in Delhi-
NCR and 360 responses were received. Out of them, 5 questionnaires were found to be incomplete
in some way or other, so, finally 355 responses were used for final analysis. Most of the
questionnaires were completed on One-on-one basis. This is why the survey yielded about 95 per
cent response rate. Though this is not perfect comparable to the 100 per cent response rate in a
study of affluent investors by Capon et al. (1994), but comparable to 98% response rate reported by
Al-Tamimi (2005), the 95% response of Chandra (2009), and 80% response in a study of similar
stature by Kumar and Chandra (2007).
Table 1 provides the summary of descriptive statistics of demographic profile of the respondents to
the survey. An overview of the summary statistics gives an idea of the characteristics of the
individual investors surveyed for the purpose of this study. The individual investors are on average
38 years old, with median age of 36 years. When compared these figures with an earlier study
conducted by the researcher (that study was also conducted in NCR region during September 2007-
February 2008; Chandra, 2009), that study also reported the mean age of 38 years. These numbers
are in contrast with the findings of another study by Graham et al. (2004) where average age was 49
years and Barber and Odean (2001) where average age was 50 years. With these numbers, it can be
said that Indian individual investors are comparatively younger to their counterparts in the USA.
Average annual income and average annual investment of the sample investors are Rs. 10,99,718
(with a median income of Rs. 12,00,000) and Rs. 5,47,605 (with a median investment of Rs.
6,00,000) and comparable to the results reported in the previous study (Chandra, 2009), where
median income was Rs. 8,00,000. That study did not report about investment numbers. The
[13]
investors surveyed in the present study were well educated as more than 21 percent had
postgraduate education and more than 4o percent were professionally qualified. As far as gender of
the respondents is concerned, about 64 per cent of the surveyed investors were male, and remaining
36 per cent of the sample respondents were female. This indicates that female’s participation in
investment front is still way below comparable to their male counterparts. This may also be
explained by the fact that female investors do not feel comfortable in managing money, though this
is not yet statistically proved by any research. This is just an observation. Though Barber and Odean
(2001) reported an interesting finding that women are very risk averse while making trading
decisions, but men frequently suffle their investment portfolio and make some unwarranted errors,
thereby causing monetary losses on their investments.
Univariate Analysis
From the summary statistics communicated in the Table 2, the survey data does not remark any
typical behaviour that appears significant among Indian individual investor. In fact each variable
covered in the survey captures one of the psychological biases identified in the Indian context.
In what follows the researchers look individually for each psychological attitude considered in the
survey questionnaire.
Use of Heuristics: Heuristics by definition are the process by which people reach conclusions,
usually by what they find for themselves, from available information. This often leads them to
develop the thumb rules, but these are not always accurate. These heuristics cause investors commit
errors in particular situations. A review of the heuristics identified and tested with the help of the
survey questionnaire is certainly helpful in understanding the initial patterns of individual investor
behaviour in Indian stock market. A total of five heuristics has been appreciated in the questionnaire
by 7 variables. E ach heuristics and its influence on the sample investors will be discussed as
following:
E xploring the ‘Representativeness Bias’ Influence: The analysis provides with asymmetric
results in respect of the representativeness bias. That is the results show strong evidence for
existence of representativeness bias in some aspects and strong evidence for the absence of this
psychological bias among Indian individual equity investors. The representativeness bias is
appreciated in the questionnaire by V1 and V2. Results show that the sample investors seem to be
following the performance of a stock in the recent past. Majority of surveyed investors appreciate
the concept of ‘hot’ stock on the basis of the recent past performance of such stocks. That is the
sample investors are likely to consider the recent past performance of any stock at best representing
its near future return, hence, found it worth investing in such stocks. Remaining of the surveyed
investors seems to be ignoring the recent past performance of any stock as their decision-making
criteria.
[14]
E xploring the ‘Overconfidence Bias’ Influence: It is said that investors are vulnerable to the
overconfidence barrier. They tend to have too much confidence in the accuracy of their own
judgments. As they find out more about a situation, the accuracy of their judgment may well not
increase, but their confidence will, as they equate quantity with quality. Confidence also tends to
increase if they are given incentives to perform well. When participating investors were asked
whether they use their predictive skills deemed to have obtained from their experience from their
investment portfolios, in order to time and outperform the market. The overconfidence tendency
seems not to be popular among the respondent investors.
E xploring the ‘Anchoring Bias’ Influence: Anchoring heuristics refers to individuals’ tendency to
base estimates and decisions on known ‘anchors’ or familiar positions, with an adjustment relative to
this starting point. They are better at relative think ing as compared to absolute think ing. This heuristics is
found significantly among individual investors. During the survey of individual investors, they were
asked whether their trading was influenced by recent experiences about price in the market, and
whether they use the purchase price of stocks as a reference point while making trading decisions.
Results indicate that investors are more likely to react in the defined manner and they are very much
likely to be influenced by their recent experiences; the trading decisions of about 53.2 per cent
sample investors is well influenced by their price experiences in the recent past. Remaining of the
investors i.e. 46.8 per cent investors were most unlikely to be influenced by past prices while making
trading decisions. On the other hand, it is also observed that large portion of surveyed investors
were suffering from anchoring bias, as they used their purchase price as a reference point for their
trading decisions.
E xploring the ‘Gambler’s Fallacy’ E ffect: This heuristics is appreciated in the questionnaire by
the variable V 6 and the results for this variable support the overconfidence bias influence among the
surveyed investors. For the purpose of survey, respondents were asked whether they would be able
to anticipate the end of good or poor returns at the stock market. It is worth noticing that this
question was specially referring to the anticipation of market returns. This relatively poor score
shown in univariate statistics indicates those investors are least likely to anticipate about stock
market returns. The frequency statistics also show that only 35.5 per cent of respondents were of
view that they could successfully anticipate the market returns whether it would be good or poor at
the market index. The value is very much similar to that of the variable capturing the overconfidence
bias, where investors were asked whether they could predict the timing to outperform the market. It
provides the evidence that less number of investors surveyed were seemingly able to anticipate the
market returns.
E xploring the ‘Availability Bias’ Influence: Investors are more likely to act on this readily
available information. Growth stock is a very hot story and everyone likes a stock a stock that goes
up very fast. Individuals as well as Institutional investors fall prey to this availability heuristics, start
believing in the growth story and chase growth stocks. This availability heuristics comes into play
[15]
while individuals taking trading decisions in the markets. Survey results show that sample investors
do not hold it good for their investment decision-making. The above-average-score of this factor
shows that they are likely to be swayed away by media reports and join the bandwagon. And
frequency statistics also support this finding that about 57.7 per cent individual investors seem to be
taking their investment/trading decisions on the basis of the readily available information about
their choices. This figure leads us to believe that more investors are swayed away by the vividly
publicized information about any stock and choose to buy that stock above those with less media
attention.
Use of the Prospect Theory: The Prospect Theory is one of the pillars upon which the much of
the behavioural economics rests. This theory was proposed by Kahneman and Tversky (1979) while
expounding the risk theory. This section captures the evidence of the existence of the psychological
biases among the sample investors.
E xploring the ‘Loss Aversion Bias’ Influence: The results of the analysis show the asymmetric
evidence for the loss aversion bias. The numbers give strong evidence for investors being risk averse
when faced with sure gains and weak evidence for investors being risk takers when faced with sure
loss. Investors tend to react to the stock markets under the grip of greed and fear. They become
greedy in a bullish market, but become fearful when the market is falling. The results connote that
sample investors are likely to be risk averse when faced with sure profits. About 55 per cent of
sample respondents seem to become risk averse when they are faced with a sure profit. On the other
hand, the variable V9 is appreciated in the questionnaire to capture the risk-seeking nature of
investors. The mean value of this variable is 2.606 (S.D. = 1.023); it provides evidence for the
likelihood of risk-seeking attitude when faced with sure loss. Summary statistics shows that only 48.6
per cent of sample respondents seem to be risk-seeking when faced with sure loss. According to
these data, the researcher believes that most of the sample investors (about 55 per cent) chose to
become risk averse when faced with sure gains, but lesser of them (only 48 per cent around), on the
contrary, tend to become risk-seeker in case faced with sure loss.
E xploring the ‘Regret Aversion Bias’ Influence: As per the Prospect Theory, losses are three
times more painful as compared to the pleasure experienced by a gain of similar magnitude. What is
the reason behind this anomaly? The answer is that people hate regret and losses produce regret.
Sometimes, with the sole view of insuring themselves against future losses and subsequent regret,
people weigh only the negatives and let the potential of regret influence their decision-making
process. The univariate statistics here show some evidence for the influence of this contextual
behavioural tendency on Indian individual investors. This behavioural tendency is appreciated in the
survey questionnaire by the variables V 10 and V11. The variable V10 traduces individual investors’
tendency to delay in selling stocks that have decreased in value. The mean for this variable is 2.910
(with standard deviation = 1.139) which indicates that most of the sample investors avoid selling the
stocks that have decreased in value. This is somewhat gratification of losses. They believe the prices
to recover and delay in selling such stocks till the prices bounce back to their reference point (see the
[16]
variable V5: the Anchoring Bias). This view is supported by the data on variable V11 (Mean score =
3.406, S.D. = 0.802). This variable traduces investors’ tendency to book profits on the stocks that
have increased in value even if the prospect of its gaining higher value in near future is significant.
E xploring the ‘Mental Accounting Bias’ Influence: The mental accounting error is appreciated
in the questionnaire by the variable V 12. This variable V 12 traduces individuals’ tendency to treat each
element in their investment portfolio separately. Sample respondents were asked whether they
would treat it as overall loss if two out of ten stocks in their investment portfolio are reduced in
their value significantly. The mean score of this variable 2.80 (S.D. = 0.603) indicates that individual
investors seem to taking the losses separately, not as overall loss on the entire portfolio. Most of the
respondent investors consider loss on stocks separately rather calculating it as loss on the portfolio.
Frequency statistics for this variable shows that 64.2 per cent of sample respondents believe that if
any stock in their portfolio has depreciated in its value, this loss must be attributed to that asset only
and it cannot be adjusted against the value of the entire portfolio. The data from this survey
provides strong evidence for mental accounting effect on Indian individual investors. This can be
explained by the observation that lack of proper awareness and access to right source of information
make them more conservative and hence they get influenced by mental accounting error. Though
this trend is not tested by statistics, it is the observation made by the researcher during the survey.
E xploring the ‘Contextual Factors’ Influence: The present study also examines the impact of
contextual factors in addition to the psychological factors, on individual investors behaviour.
Psychological factors were identified and extracted from psychological experiments and available
literature, while contextual factors were taken into account after interactions with industry
professionals. They suggested five such contextual factors that seem to be having influence on
investor behaviour in stock markets. These contextual factors are: mark et share and reputation of the firm
(V13), accounting and financial information (V14), publically available information through various media (V15),
advocate recommendations including those from brok ers, family and friends (V16), and personal financial needs (V17).
While collecting data through questionnaires, sample respondents were asked to rate the likelihood
of impact of these listed factors while making investment/trading decisions. The rating scale for
these five factors was five-point Likert scale, 1 being the least likely to influence the decisions and 5
being the most likely to influence their decisions.
The data from the survey gives asymmetric evidence for these contextual factors; it gives poor
evidence for first two listed factors (i.e., market share and firm’s reputation, and accounting and
financial information), but provides strong evidence for last three factors, namely, publicly available
information through various media, advocate recommendations and personal financial needs. The
mean score for the first factor, market share and firm reputation is 2.406 (S.D. = 0.802); it means
sample investors rate it with moderate influence on their decision-making in the stock market.
Second factor accounting and financial information has a mean score of 2.701 (S.D. = 0.458) which
is again very close to the first factor. It connotes that investors seem to be giving little-above-average
importance to the accounting and financial information. Other three factors, namely, publicly
[17]
available information through various media, personal needs, and advocate recommendations have
significant mean scores of 3.299 (S.D. = 0.643), 3.597 (S.D. = 0.491), and 4.000 (S.D. = 0.774)
respectively. It can be said that sample investors are giving significant weight to all factors except
market share and firm’s reputation, and accounting and financial information.
The observations of five contextual factors’ influence on investor behaviour give asymmetric results
wherein investors seem to be giving more importance to some factors than to others. It is seen that
personal needs and advocate recommendations are highly rated by investors as compared to market
share and accounting information. The results from these observations are not concluding about the
significance of the factors influencing investor behaviour in Indian stock market. It is, therefore,
important to run more sophisticated analytical methods in order to find a crystal clear picture of
factors influencing individual investor behaviour in Indian stock market.
Multivariate Analysis
With an objective to determine the suitability of data for factor analysis, the Kaiser-Meyer-Olkin
Measure of Sampling Adequacy (MSA) and the Bartlett’s Test of Sphericity were applied. The
Kaiser-Meyer-Olkin (KMO) Measure of Sampling Adequacy is a statistic that indicates the
proportion of variance in the variables that might be caused by the reduced factors. Kaiser (1974)
recommends that a bare minimum of 0.5 and that values between0.5 and 0.7 are mediocre, values
between 0.7 and 0.8 are good, values between 0.8 and 0.9 are great and values above 0.9 are superb
(See Hutcheson & Sofroniou, 1999, pp. 224-225). The results from these tests are given in Table 3:
Table 3: KMO and Bartlett’s Tests
Kaiser-Meyer-Olkin Measure of Sampling 0.932
Adequacy
Approx. Chi-Square 13292.550
Bartlett’s Test of Sphericity df 136.000
Sig. 0.000
High value of KMO (0.932) indicates that a factor analysis is quite useful for the data being used in
this study. The KMO figures provide strong evidence for sampling adequacy for these data.
Similarly, the significant value for Bartlett’s test of sphericity is 0.000 which indicates that there exist
significant relationships among variables. The output of KMO and Bartlett’s tests supports the view
that factor analysis is very much useful for the present data.
The result of multivariate analysis is mainly focused on combining all the psychological and
behavioural factors considered in the survey instrument in order to resume them and to reduce them
into few behavioural axes that should describe and lead the Indian individual investor behaviour.
E ach axe will regroup all the correlated psychological and behavioural factors. To achieve this
multivariate analysis, the Principal Components Analysis was run on the data collected through the
survey of sample investors.
[18]
Determining the Number of Meaningful Components to Retain: The PCA approach suggests
that the number of components extracted is equal to the number of variables being analyzed,
necessitating that it is to be decided just how many of these components are truly meaningful and
worthy of being retained for rotation and interpretation. In general, it is expected that only the first
few components will account for meaningful amounts of variance, and that the later components
will tend to account for only trivial variance. The next step of the analysis is, therefore, to determine
how many meaningful components should be retained for interpretation. Among others, following
three criteria may be used for the purpose (Cattell, 1966, Stevens, 1986): the eigen value-one criterion, the
scree test, and the proportion of variance accounted for. First, in the Principal Components Analysis, the
eigen value-one criterion, also known as the Kaiser criterion is one of the most commonly used
criteria for solving the number-of-components problem. With this approach, any component with
an eigen value greater than 1.00 is retained for rotation and interpretation. Second, with the scree
test (Cattell, 1966), the eigen values associated with each component are plotted and observed for a
“break” between the components with relatively large eigen value and those with small eigen values.
The components that appear before the break are assumed to be meaningful and are retained for
rotation; those appearing after the break are assumed to be unimportant and are not retained. A third
criterion in solving the number-of-factors problem involves retaining a component if it accounts for
a specified proportion (or percentage) of variance in the data set. An alternative criterion is to retain
enough components so that the cumulative percent of variance accounted for is equal to some
significant value. When the ‘cumulative percent of variance accounted for’ is used as a criterion for
solving the number-of-components problem, it is suggested to retain enough components so that
the cumulative percent of variance accounted for at least 70% (and sometimes 80%).
Following the above mentioned criteria for solving the number-of-components problem in this
study, the components matrix is formed for further orthogonal rotation using Varimax rotation
algorithm which is standard rotation method (Kaiser, 1958). The multivariate analysis extracts
obviously 17 behavioural components, but only five components were retained for rotation and
interpretation, as these five components were judged sufficient to explain the significant data
variance and also qualified the above mentioned criteria for solving the number-of-components
problem. In fact, all the five components so selected seem to explain above 91% of total data
variance, and the remaining variance is explained by other variables as given in Table 4:
Table 4: Components and Variance E xplained
Sr. No. Components E igen Value % of Variance Cumulative Variance
E xplained
1 Component 1 7.351 43.243 43.243
2 Component 2 2.918 17.163 60.406
3 Component 3 2.184 12.845 73.251
4 Component 4 1.703 10.015 83.266
5 Component 5 1.318 7.754 91.020
[19]
An observation of the table above provides an insight that only these five components extracted
from the Principal Components Analysis are significant enough to retain for rotation and further
interpretation as all these components qualified the criteria of the eigen value-one, the variance
accounted for (or the cumulative variance above 80%), and also the scree plot break. As can be seen,
the variance proportion explained begins to decrease from sixth component onward (figure 1).
Moreover, the correlation coefficients of these pertinent five factors or components with the initial
variables are higher than 0.5.
The Behavioural Factors Defined: According to the extracted coefficients these five pertinent
behavioural components were named as follows: prudence and precaution attitude, conservatism, under
confidence, informational asymmetry, and financial addiction. The total variance accounted for, by all the five
factors with eigen value greater than 1 is 91.02% which sufficiently significant, and the remaining
variance is explained by the other variables. Among the five factors, the first factor accounts for
around 43.24% of variance which is the prime factor influencing investment behaviour of any
Indian individual investor. The detailed values obtained from the PCA tests for these five factors
with labels, factor loadings and communalities for various statements included as variables are given
in the Tables 5-10.
[Insert Tables]
Following are the detailed analysis for each of the five components extracted from the Principal
Components analysis:
[20]
Prudence and Precaution Attitude: The first component is prudence and precaution attitude as
the statements or variables included under this component are related to it. This component is an
important factor because it accounts for more than 43% of data variance. E ven after rotation, this
component represented by these variables accounts for 29.5% of data variance. E ach variable
included under this factor is associated with a different kind of behavioural attitude. These variables
underline the symmetric behavioural attitude of risk aversion and calculated trading decisions. So, in
summary this behavioural factor traduces the prudent and cautious attitude of Indian individual
investors in the stock market. In fact the Indian individual investors tend to use trend analysis for
their trading decisions. Their decisions are also based on the firm’s market share and reputation; but
they also tend to use their purchase price as reference point (which provides sufficient evidence for
the existence of the prospect theory). Although they showed risk seeking attitude in certain loss-
making circumstances, they tend not to hesitate in profit booking whenever such opportunities arise.
In certain instances, they would prefer to retain the losers in their portfolios. This prudence attitude
shown by the surveyed investors may be justified by an excess risk aversion and insufficient
familiarization with the stock market which might frighten them at times.
Conservatism: According to the multivariate analysis results, the sample investors seem to be
conservative rather than adaptive. This behavioural component accounts for about 17.16% of data
variance explained. The rotation sums of squared loadings show that this component accounts for
about 21.45% of data variance after orthogonal rotation of component matrix. It is, therefore,
evident that this component is second most influential factor/component in case of Indian
individual investor behaviour in stock market. The underlying variables underline the investor
behavioural tendency to be attached to the past data and/or events and also the traditional
approaches to take trading decisions by a layman investor. Investors under the influence of this
behavioural factor/component tend to be risk averse most of the time, and derive their trading
decisions based on what is recommended by their acquaintances and their past trading experiences
in the market. One of the factors which affect them is their personal financial needs. Results indicate
that they are not confident enough to time and outperform the market using their predictive skills
and they rarely bother about the colourful stocks (i.e. the stock which are much talked-about and
presented with flying colours in the investment circle). So, Indian individual investors seem to be
significantly influenced by the conservatism psychological bias.
Under Confidence: The third component extracted by the analysis is attributed as the under
confidence and the reference variables underline the individual investors’ tendency to react promptly to
whatever they come across with; they don’t tend to analyze their decisions and easily change their
positions. They tend to revise quickly their decisions based on their immediate past experiences.
Also they are much influenced by others’ say on their considerations. This component is explained
by 12.95% of data variance before rotation and about 15.52% of data variance after rotation, thereby
making the result more reliable. The set of variables covered under this component traduces that
individual investors in India take their trading decisions under the influence of their recent past
experience. They place high importance to the stock which is highly promoted and much-hyped,
[21]
also considered as ‘hot stock’, rather objectively evaluating the alternative. So, under confidence
affects their trading decisions to the extent that they change easily their positions and quickly revise
their decisions as soon as they see any new ‘hot stock’; this reflect their under confidence in their
own decisions and ability. It can, therefore, be said that individual investors in India are influence by
the under confidence psychological bias.
Informational Asymmetry: The next important component derived from the multivariate analysis
is named as informational asymmetry. This component explains the asymmetric pattern of distribution
and usage of information among the sample investors. The component accounts for about 10% of
data variance before rotation and about 12.5% of data variance after Varimax rotation of variables.
The set of variables explaining this component leads to infer that individual investors are suffered
from informational inferiority complex. They tend to rely heavily on the easily available and
accessible information. They are influenced by the information hovering around them and which
can be easily used by investors for their decisions; rather they don’t tend to check the reliability of
the information and prefer to those piece of information which are easy to incorporate into their
decisions. Instead of incorporating all the publically available information as suggested by the
standard definition of market efficiency theory, investors tend to discount the information that
seems complex to incorporate into their decision-making process, and adopts only those easily
available and adjustable. They don’t practice information mapping whereby information is classified
according to the sources and their reliability, and then being considered for decision-making on top-
down basis. Though Indian individual investors are seen using different sources for their
informational needs, they lack the objectivity in rationally using the appropriate ones. So, they are
influenced by the informational asymmetry psychological bias.
Financial Addiction: The fifth component extracted by the factor analysis is named financial
addiction. This is so because the variables capturing this component seem to indicating that
accounting and financial information relating to the stocks concerned and their past performance
also has its bearing to the individual investors to some extent. Though this is the weakest of the five
components extracted by the multivariate analysis, the contribution of this component to investor
behaviour cannot be ignored as this component accounts for 7.75% of data variance before rotation
and about 12% after orthogonal rotation of variables. This component strongly indicates that
individual investors rely on accounting and financial statements for their trading decisions. Although
the accounting and financial statements are said to be the part of informational sources, it was noted
earlier that investors tend to discount complex information at first instance and majority of investors
surveyed do not consider financial statements as their primary source of information. This avoidance
can be explained by the individual investors facing complexity of interpreting the financial
statements prefer to those sources of information which are easily adjustable into their trading
decisions. They tend to treat accounting and financial statements as separate piece of information.
They place less importance to this factor as compared to other sources of information among other
factors. The accounting and financial information is influencing the behaviour of individual
investors in Indian stock market.
[22]
SUMMARY & CONCLUSION
In the present paper, principal components analysis is carried out on the data collected from survey
of sample individual investors, to extract the factors influencing Indian individual investor behaviour
in stock market. E specially the psychological biases which may drive their trading behaviour were
identified. The paper started with the twin objectives of identification and verification of the
psychological biases considered to drive a momentum effect by influencing investor behaviour in
stock market.
In fact a wide array of behavioural financial literature was scanned in order to extrapolate the
psychological and contextual factors influencing individual investor trading behaviour. The results
and conclusions resorted from the behavioural financial literature and noted psychologists’
experiments as well as from the discussions with stock market practitioners provided with thirteen
psychological and contextual biases captured by seventeen variables. Then it tried to explore the
investor trading behaviour by directly addressing to the investor. To fulfill this objective the
questionnaire technique was used. The questionnaire comprised of scenario based questions relating
to all psychological and contextual biases.
The present paper deals with the data collected by survey of sample respondents based across Delhi-
NCR. The collected survey responses were then put to univariate and multivariate analysis. The
principal components analysis technique method was primarily used for multivariate analysis of data
collected. The suitability of the techniques adopted in this study was tested through various
statistical tests such as KMO test of Sampling Adequacy and Bartlett’s Test of Sphericity. The results
of the principal components reveal the five underlying psychological axes that appear driving the
Indian individual investor behaviour. These five pertinent axes on the basis of the underlying
variables are named as prudence and precautious attitude, conservatism, under confidence, informational
asymmetry, and financial addiction. The results reveal some psychological axes, such as conservatism and
under confidence, which are consistent with the prior literature to some extent; but there are some
contrary behavioural axes reported by the multivariate analysis such as prudence and precautious
attitude and informational asymmetry which are not yet considered in prior literature in growing
economies, particularly in Indian context. These psychological components seem to be influencing
individual investors’ trading behaviour in Indian stock market.
[23]
References
1. Al-Tamimi, Hussein A Hassan (2005), Factors Influencing Individual Investor Behaviour:
An E mpirical Study of the UAE Financial Markets, University of Sharjah work ing paper.
2. Andrade, Sandro C., Charles Chang, and Mark S. Seasholes (2005), Trading shocks, assets
prices, and the limits of arbitrage, University of California, Berk eley Work ing paper.
3. Barber, Brad M., and Terrance Odean, (2000), Trading is hazardous to your wealth: The
common stock investment performance of individual investors, Journal of Finance 55, 773-806.
4. Barber, Brad M. and Terrance Odean (2001), Boys will be Boys: Gender, Overconfidence,
and Common Stock Investment, Quarterly journal of E conomics, Vol.116, pp. 261-292.
5. Barber, Brda M., and Terrance Odean (2005), All the glitters: The effect of attention and
news on the buying behaviour of individual and institutional investors, University of California,
Davis Work ing paper.
6. Barber, Brad M., Yi-Tsung Lee, Yu-Jane Liu, and Terrance Odean (2005), Who loses from
trade? E vidence from Taiwan, University of California, Davis Work ing paper.
7. Brown, G.W. and Michael T. Cliff (2004), “Investors Sentiment and the Near-term Stock
Market”, Journal of E mpirical Finance, Vol. 11, pp. 1-27.
8. Capon, Noel, Gavan J. Fitzsimons, and Rick Weingarten (1994), Affluent Investors and
Mutual Fund Purchase, International Journal of Bank Mark eting, Vol. 12, No. 3, pp.17-25.
9. Cattell, R.B. (1966), The Scree Test for the Number of Factors, Multivariate Behavioural
Research, Issue 1, pp. 245-276.
10. Chandra, Abhijeet (2009), Individual Investor Trading Behaviour and the Competence
E ffect, The ICFA I Journal of Behavioural Finance, Vol.VI, No.1, pp.56-70.
11. Choe, Hyuk, Bong-Chan Kho, and Rene M. Stulz (1999), Do foreign investors destabilize
stock returns: Theory and evidence, Journal of Financial E conomics, 72, 485-518.
12. Coval, Joshua D., David A. Hirshleifer, and Tyler Shumway (2002), Can individual investors
beat the market? Harvard University Work ing paper.
13. Develis, Robert F. (1991). Scale Development: Theory and Applications, Sage Publications
California.
14. E pstein, M.J. (1994) “Social Disclosure and the Individual Investor”, A ccounting, A uditing and
A ccountability Journal, Vol. 4, pp. 94-109.
15. Fisher, Kenneth L., and Meir Statman (2000), “Investor Sentiment and Stock returns”,
A ssociation for Investment Management and Research, pp. 16-23.
16. Fischer, Rene, and Ralf Gerhardt (2007), “Investment Mistakes of Individual Investors and
the impact of Financial Advice”, E uropean Business School Work ing Paper.
17. Graham, John R., Harvey Campbell R. and Huang Hai (2004), Investor Competence,
Trading Frequency and Home Bias, Duk e University work ing paper.
18. Griffin, John M., Jeffery H. Harris, and Selim Topaloglu (2003), The dynamics of
institutional and individual trading, Journal of Finance, 58, 2285-2320.
19. Grinblatt, Mark, and Matti Keloharju (2000), The investment behaviour and performance of
various investor types: A study of Finland’s unique dataset, Journal of Financial E conomics, 55,
43-67.
[24]
20. Hodge, F. D. (2003), “Investors’ Perceptions of E arnings Quality, Auditor Independence,
and the Usefulness of Audited Financial Information”, A ccounting Horizons, Vol. 17, pp. 37-
48.
21. Hvidkjaer, Soeren (2005), Small trades ad the cross-section of returns, Working paper,
University of Maryland.
22. Jackson, Andrew (2003), The aggregate behaviour of individual investors, L ondon Business
School work ing paper.
23. Kadiyala, P. and R Rau (2004), “Investor Reaction to Corporate E vent Announcements:
Underreaction or Overreaction?” Journal of Business, Vol. 77, pp. 357-386.
24. H.F. Kaiser (1958): "The varimax criterion for analytic rotation in factor
analysis." Psychometrik a 23: 187–200.
25. Kehneman, Denial and Amos Tversky (1973), Availability: A Heuristics for Judging
Frequency and Probability, Cognitive Psychology, 5, 207-232.
26. Kanhneman, Daniel and Amos Tversky (1979), Prospect Theory: An Analysis of Decision
Under Risk, E conometrica, March 1979.
27. Kim, K.A., and John R. Nofsinger (2003), “The Behaviour and Performance of Individual
Investors in Japan”,
28. Krishnan, R. and D. M. Brooker (2002), “Investors Use of Analysts’ Recommendations”,
Behavioural Research in A ccounting, Vol. 14, pp. 129-158.
29. Kumar, Ravinder and Abhijeet Chandra (2007), Individual Investor Sentiment and Asset
Pricing, Synergy – the Journal of Management, Vol.9, no.2, pp.1-10.
30. Kumar, Ravinder and Abhijeet Chandra (2009), Selection Behaviour of Individual Investors:
E vidence from Mutual Fund Investment, Proceedings of the International Finance Conference, IIM
Calcutta, 3-5 December 2009.
31. Lim, Sonya Seongyeon (2004), “Do Investors Integrate Losses and Segregate Gains? Mental
Accounting and Investors Trading Decisions” DePaul University Work ing paper.
32. Malmendier, Ulrike, and D. Shanthikumar (2003) “Are Small Investors Naïve?”, Stanford
University Work ing Paper.
33. Merikas, A. and D. Prasad (2003), “Factors Influencing Greek Investor Behaviour on the
Athens Stock E xchange”, paper presented at the A nnual Meeting of the A cademy of Financial
Services, Denver (Colorado, USA), October 2003.
34. Nagy, R. A. and R. W. Obenberger (1994) “Factors Influencing Investors Behaviour”,
Financial A nalyst Journal, Vol. 50, pp. 63-68.
35. Nunnally, C. J. (1976), Psychometric Theory, McGraw Hill New York .
36. Odean, Terrance (1998), Are investors reluctant to realize their losses? Journal of Finance, 53,
1775-1798.
37. Odean, Terrance (1999), Do investors trade too much? A merican E conomic Review, 89, 1279-
1298.
38. Richards, Anthony (2005), Big fish in small ponds: The trading behaviour of foreign
investors in Asian emerging equity markets, Journal of F inancial and Quantitative A nalysis, 40, 1-
27.
[25]
39. San, Granit (2005), Who gains more by trading – individuals or institutions? Working paper,
Tel Aviv University.
40. Selltiz, C., L.S. Wrightsman and W. Cook (1976), Research Methods in Social Relations¸
Holt, Rinehart and Winston, New York .
41. Shlens, Jonathon (2009). A tutorial on principal components analysis.
(http://www.snl.salk.edu/~ shlens/pca.pdf accecced on February 14, 2011.)
42. Stevens, J. (1986), A pplied Multivariate Statistics for the Social Sciences, Hillsdale, NJ: Lawrence
E rlbaum Associates.
43. Zoghalami, Faten, and Hamadi Matoussi (2009), A Survey of Tunisian Investors, International
Research Journal of Finance and E conomics, Issue 31, pp. 66-81.
[26]
Table 2: Descriptive Statistics
Variables Mean Std. Deviation Analysis N
V1 3.70 .782 355
V2 2.70 1.006 355
V3 1.81 .600 355
V4 3.50 .501 355
V5 4.20 .600 355
V6 1.60 .666 355
V7 2.70 .458 355
V8 2.90 .947 355
V9 2.61 1.023 355
V10 2.91 1.199 355
V11 3.41 .802 355
V12 2.80 .603 355
V13 2.41 .802 355
V14 2.70 .458 355
V15 3.30 .643 355
V16 3.60 .491 355
V17 4.00 .774 355
[27]
Table 5: Total Variance E xplained
Initial E igen Values E xtraction Sums of Squared Loadings Rotation Sums of Squared Loadings
Component Total % of Variance Cumulative % Total % of Cumulative Total % of Cumulative
Variance % Variance %
1 7.351 43.243 43.243 7.351 43.243 43.243 5.016 29.508 29.508
2 2.918 17.163 60.406 2.918 17.163 60.406 3.647 21.454 50.962
3 2.184 12.845 73.251 2.184 12.845 73.251 2.640 15.528 66.490
4 1.703 10.016 83.266 1.703 10.015 83.266 2.128 12.521 79.010
5 1.318 7.754 91.020 1.3 7.754 91.020 2.042 12.010 91.020
6 .737 4.334 95.354
7 .471 2.768 98.122
8 .319 1.878 100.000
9 6.297E -15 3.704-14 100.000
10 3.883E -15 2.284E -14 100.000
11 3.170E -15 1.865E -14 100.000
12 1.067E -15 6.278E -15 100.000
13 -8.537E -16 -5.022E -15 100.000
14 -2.114E -15 -1.244E -14 100.000
15 -4.532E -15 -2.666E -14 100.000
16 -6.819E -15 -4.011E -14 100.000
17 -1.117E -14 -6.572E -14 100.000
E xtraction Method: Principal Component Analysis
[28]
Table 6: Component Matrixa
Component
Variables 1 2 3 4 5 Table 7: Rotated Component Matrixa
V2 .932 Component
Variables 1 2 3 4 5
V13 .000
V6 .876
V5 .846
V10 .841
V6 .816
V13 .822 .438
V11 .780 .171
V9 .783
V9 .735
V11 .731 .456
V3 .729 -.643
V3 .702 .512
V8 .672 .638
V2 .645 .663
V10 .640 -.629
V8 .869
V12 .624 -.610
V16 .792 -.491
V1 .606 .612 .119
V15 .773
V15 .558 .435 .422
V17 .769 -.492
V16 .927
V7 .984
V17 .500 .500
V5 .605 .673
V4 .729
V14 .949
V7 -.514 .467 .635
V4 .946
V14 .480 .703
E xtraction Method: Principal Component Analysis
V1 .540 .622
a. 5 components extracted. V12 .412 .408 .482 -.519
E xtraction Method: Principal Component Analysis
Rotation Method: Varimax with Kaisar Normalization
a. Rotation converged in 8 iterations
[29]
Table 8: Communalities Table 9 – Component Transformation
Variables Initial E xtraction
Matrix
[30]