INTRODUCTION
Research background
The Vietnamese stock market was established in 1998 It consists of the Ho Chi Minh City stock exchange (HOSE) and the Ha Noi stock market (HAS) At the first stage, in 2000, the Vietnam stock market had only 2 listed companies and 4 security companies However, after more than ten years, there are currently 682 member securities companies These mark the development of the Vietnamese stock market in general and the HOSE specifically Starting at 100 points in July 2000, after one year in operation in June, 2001, the VN-Index had increased fivefold and reached a peak at 571 points Investors were overly excited and dreamed of earning quick money as demand rose significantly even though there were only a few listed stocks on the stock exchange However, it was lack of knowledge and trading experience of investors as well as the insufficient support from the authorities that made the VN- Index to suddenly unceasingly fall until it reached the bottom at 139 points in March
2003 (Huy, 2010) Investors who joined the market during this period and who were not able to jump out quickly had to face severe financial difficulties due to the huge loss of assets The stock market then seemed to go into a hibernative status until 2005 and finally “woke up” in 2006 The boom started in the second half of 2006 and the stock market rocketed up to 1170 points by March 2007 It fluctuated around the 1000 point mark until October, 2007 The Ho Chi Minh stock market had never been
“hotter” than at that time However, VN-Index went again into a decline stage after being pushed up to the peak The Ho Chi Minh stock market experienced a gloomy year in 2008 and VN-Index only stopped decreasing in February 2009 at 245 points
Economic experts and analysis explained the sharp decline of the VN-Index that was affected by various factors such as the tightening of monetary policies, especially the lending for stock investment, high deposit interest rates, high inflation rates, and a recession in the United States economy Lack of timely intervention by the authorities was also a reason why VN-Index fell so dramatically (Vo and Pham, 2008, p.15)
From 2009 to the first quarter of 2011, VN-Index continued to undergo many ups- and-downs: reaching another peak at 542 points in May 2010, another bottom at 351 points in November, 2010 However, it seemed to fluctuate around 400-500 points and no significant amplitude was found in 2012
Table 1: The highest and lowest VN-index from 2000 to 2012
Year Highest prices Lowest prices
Problem of statement
During twelve years, the HOSE specifically and the Vietnamese Stock exchange in general went through many different stages Its prices fluctuated, up and down, and it seemed to be difficult for investors to make sound investment decisions
Especially, with the boom of VN-Index 2007, the Vietnam Stock Market was negatively impacted by foreign newspapers The Financial Times showed that the HOSE was a „too hot place‟ and overstated its value The AMCHAM (The American Chamber of Commerce) declared that the Vietnam stock market‟s growth was impractical, just another case of a „bubble market‟
The Efficient Markets Hypothesis (EMH) could not explain causes leading to this bubble market because EMH is referred to as “the notion that stocks already reflect all available information” (Bodie, Kane & Marcus, 2008, p 359) and capital markets are informationally efficient (Fama, 1998, p 383 – 417) Meanwhile, behavioral finance assumes that, in some circumstances, financial markets are informationally inefficient (Ritter, 2003, p 429 – 437) and factors of behavioral finance could be considered in the situation of bubble market (Waweru et al, 2008, p.25)
Behavioral finance can be helpful in the bubble market because it is based on the psychology to explain why people buy or sell stocks (Waweru et al., 2008, p.25)
Behavioral factors include overconfidence, representativeness, availability, herding, loss aversion, regret aversion, gambler‟s fallacy, over-underreaction, etc (Ritter, 2003, p 437) Many researchers consider behavioral finance as a good theory to understand and explain feelings and cognitive errors affecting investment decision-making (Waweru et al., 2008, p.25) Supporters of behavioral finance believe that the study of social sciences such as psychology can help to reveal the behaviors of stock markets, market bubbles and crashes As shown below:
In Europe, Johnsson, Andrén, Lindblom & Platan (2002) reported results of behavioral factors of individual investor and institutional investors in Sweden that positively impact on their investment decision It is accepted that 67% of the total amount of individual investors are influenced by the factor of loss aversion bias, 33% of the total number of investors are impacted by the factor of representative bias, 32% of total amount of investors are controlled by the factor of regret aversion bias, etc
Besides European countries, behavioral finance is also discovered in Asia
Chandra, Abhiject and Kumar, Ravinder (2011) research determinants of individual investor behavior in the Indian stock market that have influence on their investment decisions and results as follows: Representativeness bias with strong existence, Overconfidence bias that investors tend to have too much confidence in the accuracy of their own judgments, 53,2% of total number of investors are controlled by the factor of anchoring, 35,5% of total amount of participants are affected by the factor of gambler‟s fallacy on their investment decision 57,7% of total amount of investors are impacted by the factor of availability bias And 55% of total numbers of investors are influenced by the factor of loss aversion bias on their decision, etc
In addition, Vuong & Dao (2012) discover individual investors‟ behavioral biases in the Vietnamese stock market; namely, there are 51,2% of the total number of investors are influenced by the factor of regret aversion bias on their investment decision 47,7% of the total amount of investors are impacted by the factor of loss aversion bias on their investment decision Similarly, the factor of anchoring bias:43%, the factor of overconfidence bias (42,4%) the factor of self-control bias (42.4%), illusion of control bias (41.3%), confirmation bias (41.3%), framing bias (34.9%), conservatism bias with 30.3%, representative bias with 18% and 17.4% of the total amount of investors are affected by the factor of availability bias
There are two reasons why behavioral finance is important and interesting when applied to the Vietnamese stock market - HOSE First, behavioral finance is still a new topic for study Until recently, it is accepted as a feasible model to explain how investors of financial markets make decisions and then how these decisions influence the financial markets (Kim and Nofsinger, 2008, p.1) Secondly, due to some evidence such as subjective, academic, and experimental, it is concluded that Asian investors, included are the Vietnamese, usually suffer from cognitive biases more than people from other cultures (Kim and Nofsinger, 2008, p.1) Therefore, the consideration of the factors influencing the Vietnamese investors‟ decision-making process cannot ignore the behavioral elements This study hopes to enrich the number of studies using behavioral finance for making investment decisions by Vietnamese investors in the stock market.
Research questions
The paper identifies behavioral factors influencing investment decision- making of individual investors at the HOSE Additionaly, the paper also identifies impact levels of behavioral factors on the investment decision-making of individual investors at the HOSE The paper designs research questions as follows:
1/ What are the behavioral factors influencing individual investors‟ decisions at the HOSE?
2/ How strong is impact of behavioral factors on investment decision making of individual investors at the HOSE?
Research scope
There are only two stock exchanges in Vietnam; one is in the Ha Noi Capital and one is in Ho Chi Minh City This study focuses on the Ho Chi Minh stock exchange because Ho Chi Minh City is the biggest and most developing city in Vietnam The Ho Chi Minh stock market has become the yardstick of the economy‟s wealth and helps enterprises and investors to raise capital for their business enterprises Moreover, due to time limitations, choosing HOSE for this research will help the author shorten the distance, provide more accurate survey information and offer more effective interview data as well.
Research method
The paper employs both qualitative and quantitative methods This research process involves pilot study and official research This pilot study is based on the qualitative method and pilot interview The qualitative method helps the author define and discover the behavioral factors of investors affecting their investment decision- making in order to adjust and complement this official research The pilot interview that is performed by the direct interview approach aims to make questionnaires more perfect before the official research The author directly interviews 10 investors who have more than 10 years of experience with stock investments The quantitative research is carried out by survey questionnaires that are sent to 220 investors who have invested in the HOSE First, the author sends the questionnaires to 52 investors (pilot survey) in order to test the reliability of the measurement scales After deleting some questions that appear to not be relevant or reliable, the author sends the survey questionnaire to 170 more investors
In this study, the hypotheses can be built based on the existing behavioral finance theories, the quantitative method is used first to test these hypotheses and then the qualitative method is used to analyze the results with a deeper inspection To test the reliability of the scales measurement , the paper uses SPSS software to test Cronbach‟s Alpha based on standardized items After that, the paper tests scales measurement validity: factor analysis - EFA (Exploratory Factor Analysis) Finally, the paper applies regression for testing assumptions and hypothesis.
Significance of the research
• Help investors or organizations consider and analyze these behavioral factors before making decisions for investment
• Employ these behavioral factors effectively to increase success in business.
Structure of the study
Chapter 1, Introduction, mentions two parts The first part relates to the overview of HOSE from the initial stages of its opening to the present The second part involves the problem statement and gives the reasons to select the topic
Chapter 2, Literature Review, reviews behavioral factors affecting investment decision-making that previous researchers studied Behavioral factors include representativeness, availability bias, gambler‟s fallacy, herding, regret aversion, loss aversion, etc After that, the paper suggests studying six behavior factors such as representativeness, availability bias, herding, gambler‟s fallacy, mental accounting bias and over-underreaction In addition, the paper also proposes hypothesis and research models as well
Chapter 3, Methodology, presents research methods: qualitative and quantitative method In the qualitative method, the author interviews 10 investors about behavior factors they possess and their ideas regarding the content of the questionnaires After obtaining relevant and reliable questionnaires, the paper employs the quantitative method to survey 220 investors with five Likert scales (extremely disagree to extremely agree) Additionally, the paper also mentions research population, samples, steps of investment decision through SPSS software
Chapter 4, Findings, data collection and analysis, has three parts: part 1: results of the interview and report investors‟ idea Investors give valuable feedback for the questionnaires and present their ideas for the behavioral factors Part 2: results of the pilot survey (NR): the paper tests measurement reliability and validity (EFA) Part 3: results of official survey (N"0): the paper tests measurement reliability, validity (EFA) and regression and test six assumptions including independence of residuals, a linear relationship, homoscedasticity of residuals, no multicollinearity, no significant outliers or influential points and normality of the residuals After that, the paper reports the findings with a hypothesis
Chapter 5, Recommendation and conclusions, first, discusses how behavioral factors affect investors‟ decisions including representativeness, gambler‟s fallacy and over-under reaction Secondly, the paper mentions behavioral factors do not affect investment decision making of individual investors encompassing availability bias, herd behavior and mental accounting bias Next, the paper recommends to individual investors and limitation of the study Finally, the paper shows a conclusion.
LITERATURE REVIEW
Classical finance theory versus behavioral finance
According to Fama (1998), known as the father of efficient market hypothesis (EMH), EMH assumes that the capital market is informationally efficient (p 283) In contrast, behavioral finance assumes that, in some circumstances, financial markets are informationally inefficient (Ritter, 2003, p 430) According to EMH, although there are not all investors who are reasonable, the markets are proposed to be reasonable and make unbiased forecasts Meanwhile, behavioral finance believes that sometimes, financial markets do not have informational efficiency (Ritter, 2003, p
430) Additionally, Statman (1999) states that stock market efficiency has two meanings: first, it means that there is unsystematic method to beat the market Second, it means that stock price are reasonable, only reflect “fundamental” or “utilitarian” characteristic, such as risk, but not “psychological” or “value-expressive” characteristics, such as sentiment (p 18-27) Conversely, behavioral finance is the psychological decision process in recognition and prediction of financial markets (Talangi, 2004, p 3-25) In the case of capital markets, there are three categories: weak form efficiency, semi-strong form efficiency and strong form efficiency
According to EMH, the weak form efficiency is distinguished by the fact that the current price of a financial asset reflects all the historical financial information available on the market The semi-strong form efficiency is characterized by the fact that share prices adjust to publicly available new information very rapidly and in an unbiased manner, such that no excess returns can be earned by trading on that information In addition, strong form efficiency includes both semi-strong form efficiency and weak form efficiency Specifically, share prices reflect all information, public and private, but none of these can earn excess returns
With behavioral finance, Tversky and Kahneman (1979) who are recognized as the father of behavioral finance can be best explained in different phases by their works (p.263) Besides, Olsen (1998) defines that it is “the new paradigm of behavioral finance which seeks to replace the behaviorally incomplete theory of finance now often referred to as standard or modern finance.” (p.11) In addition, Fromlet (2001) proposes the following definition: “behavioral finance closely combines individual behavior and market phenomena and uses knowledge taken from both the psychological field and financial theory (p.65) Similarly, Olsen (1998) also mentions that behavioral finance is “focused on the application of psychological and economic principles for the improvement of financial decision making” (p.11)
In short, behavioral finance represents a revolution in financial theory It highlights the psychological edge of investment decision making process, in strong contradiction to the EMH.
Review some behavioral factors impacting on the process of investors’ decision
Many researchers studied behavioral factors affecting investment decision making such as Representativeness, Availability bias, Anchoring, Overconfidence, etc The literature review is interested in behavioral factors influencing investment decision making by individual investors as follows:
Representativeness describes to the level of resemblance that a thing owns with its parent population (DeBondt & Thaler, 1995, p.390) or the degree to what things are similar with its population (Kahneman & Tversky, 1974, p.1124) Representativeness refers to results in some biases including investors expressing too much strength on current experiences and regardless of the average long-term rate (Ritter, 2003, p.432)
A normal situation for representativeness is that investors deduce a business‟ high long-term growth rate after some quarters of going up and positively affecting investors‟ decision (Waweru et al., 2008, p.27) Additionally, representativeness also so-called “sample size neglect” that exists as investors make an effort to draw inferences from too few samples (Barberis & Thaler, 2003, p.1065) In the stock exchange, when investors find out to buy “hot” stocks instead of poor company performance, this means that representativeness is employed (Barberis & Thaler,
2003, p.1065) Particularly, in theVietnamese stock market, representative bias of individual investors was positively 18% in 2012 (Vuong & Quan, 2012, p.9).
According to Jahanzeb, Muneer & Rehman (2012), they state that availability bias is a cognitive bias that drives humans to overestimate the probabilities of the events affiliated with memorable or vivid happenings and investors place excessive weight on the most information available while making decisions (p 534)
Additionally, in stock exchanges, availability bias appears through investors‟ preference of trading local stocks They are familiar with local stocks because of easily obtaining information, although the basic rules of diversification of portfolio management are optimality (Waweru et al., 2003, p.28) In addition, Chandra, Abhiject and Kumar, Ravinder (2011) announce that 57.7% of investors were positively impacted in the Indian stock market by this factor (p.16) Meanwhile, the percentage of investors in the Vietnamese stock market who were influenced was positively only 17.4% (Vuong & Quan, 2012, p.9)
As people wrongly believe that random event is less likely to occur following an event This is incorrect due to it is related to probability, certain events will still occur in the future „Law of small numbers” (Rabin, 2002, p.775; Statman, 1999, p.20) may lead to a Gambler‟s fallacy (Barberis & Thaler, 2003, p.1065) More specification, in the stock exchanges, Gambler‟s Fallacy appears as investors estimate incorrectly the reverse points which are carefully thought as the end of good (or poor) market returns (Waweru et al., 2008, p.27) In addition, when investors are subjective to this bias, they trend to collect suboptimal replacement, simply due to it was prior selection (Kempf and Ruenzi, 2006, p.204) Moreover, according to Chandra, Abhiject and Kumar, Ravinder (2011), they researched that in the Indian stock exchange, 35.5% of investors were positively influenced by this factor (p.15)
2.2.4 Herd behavior: buying and selling decisions
According to Chaudhary (2013), the author states that a herding behavior refers to reasonable or unreasonable actions of a large group (p 88) Investors are usually based on selective information more than individual information to make investment decisions Herd behavior can help investors succeed and have a favorable result for their investment, but may also make investors fail and have bad consequences of their investment Tan, Chiang, Mason & Nelling (2008) concentrate on herding because they prove that when a stock‟s price fluctuates, viewpoints of a large group can assist them to get profit or prevent risk (p.61) In the stock exchange, buying stocks is based on „price momentum‟ without paying attention to law of supply and demand is identified as herd behavior and leads to wrong decisions (Chaudhary, 2013, p 88) Moreover, Waweru et al (2008) show that investors are positively affected by factors such as buying, selling, investment time and a number of stocks Waweru et al (2008) also present that the investment decision of an investor is affected by decisions of other investors However, the preference of herding is based on the kinds of investors For instance, according to Goodfellow, Bohl & Gebka (2009), they expose that investment decision making of individual investors trend to come after the crowds more than institutional investors (p.213)
According to Barberis & Huang (2001), they state that mental accounting refers to “the process by which people think about and evaluate their financial transactions” (p.1248) Mental accounting permits people to manage their portfolio into different items (Barberis & Thaler, 2003, p.1108; Ritter, 2003, p.431)
Additionally, Rockenbach (2004) studies empirically and proposes that the relationship between separate investment possibilities is often not made as it is valuable to arbitrage free prices (p.524) Particularly, in the Indian stock market, this factor positively affected investors‟ decision (64.2%) in 2011 (Chandra, Abhiject and Kumar, Ravinder, 2011, p3) Meanwhile, the influence of this factor in investment decision making in the Vietnamese stock exchange was positively 11% in 2012 (Vuong & Quan, 2012, p.8)
Normally, fluctuation of market information, basic principles of the underlying stocks and stock price can make over and under reaction occur This behavioral factor highly influences investment decision making of investors
Researchers including DeBondt & Thaler (1985) studying over reaction (p.804) and
Lai (2001) researching under reaction (p.215) give results that news affects investment strategies of investors, and then positively influences their investment decision Besides, according to Barberis, Andrei Shleifer and Vishny (1998), authors state that under-reaction of stock prices to news such as earnings announcements, and over-reaction of stock prices to a series of good or bad news Lakonishok, Shleifer e
Vishny (LSV) (1994) find out that investors over-react to stocks that had been carried out well in the past and are hoped to perform well in the future get lower returns than those stocks that had had poor past performance and are estimated to have a poor future performance Moreover, DeBondt and Thaler (1995) present that the investors have over- or under-reaction when stock prices fluctuate or news occurs (p 396)
Additionally, Waweru et al (2008) also report the behavioral factors that have positive impact on investors‟ decision making including price changes, market information, past trends of stocks, customer preference, over-reaction to price changes, and fundamentals of underlying stocks (p 36)
From investors‟ behavioral finance, they make investment decision that can rational or irrational and their results of investment decision will be measured by returns they obtain According to Lin and Swanson (2003), they apply three features of returns including raw returns, risk-adjusted returns, and momentum-adjusted return through five time horizons with every day, every week, every month, every quarter and every year in order to measure their results of investment (p.208) Especially, Oberlechner and Osler (2004) believe that investment return rate can objectively assess investment decision and discover positive impact level of overconfidence on the investment decision (p.1-33)
Besides three characteristics of return are used to appraise their performance, the paper emphasizes positive impact of behavioral factors on investment decision of individual investors that mentioned in each behavioral factor above According to Kyle and Wang (1977), they expose that the overconfident investors bring positive results that are higher than their expectation not just in a short time but also for a long time (p.2) Moreover, Anderson, Henker and Owen (2005) state that people who have overconfident behavior can advantage good results (p.72) and they give a conclusion that individual who over-under reacts with information of the stock market can get greater profit than other people without over-under reaction (p.71) Kim and Nofsinger (2003) also find that positive results can occur not only when investors carefully study the past performance of buying and selling behaviors, but also in the case of stock‟s price decreasing or increasing in multiple sessions (p.2) It also is important when Lin and Swanson (2003) show that investors have positive consequence in the short time and is partly controlled by short time price momentum more than by risk-loving (p.208) Investors can obtain advantages from understanding and performing momentum tactics encompassing buying stocks that they won in the past and selling stocks that they lost in the past (Waweru et al., 2008, p.25)
In summary, there are many ways to measure results of investment decision
Some researchers including Lin and Swanson (2003), Kim and Nofsinger (2003) employ the secondary data in the stock market to assess results of investors
Suggested research model
The paper studies behavioral factors of investors in HOSE, Vietnam
Therefore it is mostly based on combining the framework of researches in Asia such as Vuong & Dao (2012) in the HOSE and results of research by Chandra and Kumar
(2011) of the Indian Stock Exchange The paper mentions the Indian Stock Exchange because both Vietnam and India are in Asia Hence, the results will be more valid by gathering and comparing the level of behavioral factors of two countries in the same area This comparison helps the paper complement reasonable ideas of what behavioral factors the paper should choose to study As Table 2 below shows behavioral factor of gambler‟s fallacy, there are 53.2% of Indian investors who were affected while Vietnamese researchers have not mentioned this behavioral factor yet
Or they have not researched the factor of representativeness in the Indian stock market Meanwhile, this factor was studied in the Vietnamese stock exchange Hence, the paper will study more of these factors Moreover, there are some behavioral factors that were researched in both the Indian and Vietnamese stock markets, but the percentage of the impacted level are such large differences For example, the factor of availability bias influenced 57.7% of Indian investors, while in the Vietnamese stock market, the figure was 17.4% And the factor of mental accounting bias, there were only 11% of Vietnamese investors who were impacted, whereas 64.2% of Indian investors were affected by this factor Therefore, these factors need to be researched further by this paper
Table 2.1: Results of behavioral factors affecting investment decision making of individual investors
Behavioral factors Percentage of affected
Percentage of affected Vietnamese investors (Vuong & Dao, 2012)
Regret aversion bias Some evidences 51,2
Herd behavior: buying and selling decisions
Over and Under – reaction Not concerned Not concerned
Although Vuong and Dao (2012) show many factors affecting investment decision-making of individual investors, it still has other factors that the paper should study According to Vietnamese experts such as Ngo (2010) and Ho (2007), they state that these four main behavioral factors often occur in the Vietnamese stock exchange: overconfidence, availability bias, herding behavior and loss aversion bias These factors are presented by Vuong and Dao (2012), except the herding behavior As a result, basing on Ngo (2010)‟s ideas of herding behavior, the paper proposes to study the behavioral factor of herding And the paper gives hypothesis 1 as follows:
H1: Herding behavior positively impacts the investment decisions of individual investors at the HOSE
In addition, Ho‟s (2007) research proved that behavioral factors often happen in the Vietnamese market including availability bias, representativeness, anchoring, gambler‟s fallacy, over and under reaction, overconfidence, mental accounting and herding behavior: selling and buying decisions These behavioral factors are also showed by Vuong and Dao (2012), but gambler‟s fallacy, over and under reaction
The paper hence suggests studying two more factors including over and under- reaction and gambler‟s fallacy Hypothesis 2 and 3 are proposed below:
H2: Over and Under-reaction positively affects the investment decisions of individual investors at the HOSE
H3: Gambler’s fallacy positively influences the investment decisions of individual investors at the HOSE
Besides, in Table 2, behavioral factors such as mental accounting and representativeness affect investment decision-making with a large different percentage between Vietnamese investors and Indian investors Particularly, 64.2% of the total number of Indian participants who are investors, are affected by the factor of mental accounting; meanwhile it is 11% for Vietnamese investors The big difference in the percentages leads to the paper to check two factors again And hypothesis 4 and
H4: Mental accounting bias positively impacts the investment decisions of individual investors at the HOSE
H5: Representativeness bias positively influences the investment decisions of individual investors at the HOSE
Moreover, in the research of Vuong and Dao (2012), the factor of availability bias only holds a low percentage – 17.4% This means that investors‟ behavioral finance which is affected by this factor is insignificant Meanwhile, according to Vietnamese experts, Ngo (2010) & Ho (2007), they emphasize investors‟ psychology in Vietnam stock market is often influenced by this factor (availability bias)
Therefore the paper will study this factor more to consider whether or not this factor has much effect on investment decision-making And hypothesis 6 is presented below:
H6: Availability bias positively influences the investment decisions of individual investors at the HOSE
In short, the paper focuses on Ho Chi Minh Stock exchange Combining behavioral factors in Vuong and Dao (2012)‟s research, Vietnamese experts‟ ideas and other researchers such as Waweru et al (2008) and Chaudhary (2013), particularly
Ho (2007) & Ngo (2010), the paper would like to study the main behavioral factors affecting investment decision-making of individual investors in Ho Chi Minh stock exchange (HOSE) such as Herding behavior, over- underreaction, Gambler’s fallacy, Mental accounting bias, Representativeness and Availability bias.
Research model
Through presented above, the research mode is proposed as follows:
A chart of research model is proposed by the author
RESEARCH METHODOLOGY
Research design
The paper employed qualitative and quantitative methods The research process involved a pilot study and official research This pilot study was based on the qualitative method and pilot interview The qualitative method helped the author define and discover behavioral factors of investors affecting their investment decision making in order to adjust and complement to official research The pilot interview that was performed by a direct in-depth interview approach aimed to make questionnaires more perfect before official research The author directly interviewed 10 investors who have worked more than 10-years with experience in stock investment This interview helped the author discover mistakes and to identify questions that were not suitable with practice in order to design a more effective questionnaire for official research The questionnaires in English and Vietnamese were shown in Appendix 5.1- qualitative research design and interview questions
Official research was carried out using quantitative methods The author used selected questionnaires to survey agreement level of 220 investors in the HOSE First, the author sent questionnaire to 52 investors (pilot survey) in order to test scales measurement reliability To test scales measurement reliability (NR), the paper used
SPSS software to test Cronbach‟s Alpha based on standardized items After deleting some questionnaires that were not reliable, the author distributed the questionnaire to
170 more investors After collecting 220 questionnaires from investors, the paper tested scales measurement reliability and validity: factor analysis – EFA (N"0)
Finally, the paper used SPSS software for regression to test six assumptions
Five Ho Chi Minh Stock exchanges have the highest trading volume, selected are the Dong A (DAS), Ho Chi Minh (HSC), Saigon (SSI), FPT (FPTS) and Viet capital (BVSC) Two individuals from each stock exchange are randomly selected to interview
Define reseach problem literature review research model qualitative research
(in- depth interview, n) proposed research model and adjusted questionnaires pilot survey
(formal questionnaires, nP) test of scales measurement reliability, NP test of Cronbach's alpha survey questionnaries, (N"0) test of scales measurement reliability (N"0)
(test of Cronbach's alpha) test of scales measurement validity - factor analysis - EFA (N"0) regression alnalysis test of six assumptions discussion, recommendation and conclusion
According to Hatcher (1994), he recommends that the minimum sample size in factor analysis (EFA) is that the number of subjects should be the larger of 5 times the number of variables, or 100 (N= 5x items) The paper has 41 questions, and therefore it needs: N= 5 x 41 = 205 Sample size of the paper is 220; therefore it satisfies this condition above Additionally, Tabachnick & Fidell (2011) mentions the rules regarding the sample size of regression analysis, the sample N>R+8*k, k is the number of independent variables In the paper, N> = 52 + 8*7 = 108 Therefore, questionnaires were sent to 220 private investors in each stock exchange to receive their responses relating to the research As a result, it met this requirement.
Adjusted research model
The qualitative research shows that most investors were affected by behavioral factors Specially, representative bias, availability bias, gambler‟s fallacy and over- under reaction most influenced an investor‟s decision With the factor of representative bias, most investors were interested in blue chip stocks In recent stages, the world economy in general and Vietnam in particular, is facing difficulties in the area of finance Therefore, investors were very careful to consider their investment behavior Most of them avoided investing in stocks that had demonstrated a poor performance With the factor of availability bias, they also supported that they prefered buying local stocks rather than foreign stocks Due to the rules set up by the Vietnamese government, it was difficult for them to trade foreign stocks Besides, they would be afraid of their limited knowledge to follow the rules of these foreign countries and notice opportunities if they traded foreign stocks They therefore chose local stocks for protecting their financial investments Moreover, they prefered buying stocks that they had more knowledge of rather than investing in strange stocks that they had not traded in before In the factor of gambler‟s fallacy, because they had over
10-years of experience of investing in stocks, they tended to agree that they could guess reverse points to help make their investment decisions Most of them predicted quite accurately and were successful in their forecast Additionally, they also agreed that if the price of a stock fell in multiple sessions, they would feel compelled to buy this stock because they believed that it would be impossible to decline any further
Fortunately, most of them were lucky when they did this In the factor of over-under reaction, they also agreed that very careful consideration would be given to the price fluctuations of stocks before deciding to invest in these stocks For this present economic environment, it was necessary to carefully consider these price changes In addition, market information was very important for their investment decisions They recognized that the Vietnamese market was sensitive and needed to be reviewed quite frequently If they ignored market information, they would face many difficulties; including the loss of their investment
In short, six behavior factors that this paper proposed to research include representative bias, availability bias, gambler‟s fallacy and over-underreaction, herd behavior and mental accounting bias Most of the investors supported that the paper should research more Herd behavior and mental accounting bias were moderate level
It means they do not support but they do not reject them either As a result, the research model after qualitative research and hypotheses do not change against at first
Scales measurement design
Questionnaires are based on practical research of Waweru et al (2008) Items are adjusted and complemented to be suitable for the Vietnamese stock market There are six behavioral factors that are proposed to study including Representative bias, availability bias, herd behavior, mental accounting bias, gambler‟s fallacy and over- underreaction The summary of correction of measured questions from interviewing was showed in appendix 5.2
3.3.1 Scales measurement of Representative bias
Variable of Representative bias is symbolized of REP According to Waweru et al (2008), there are three items to measure this variable First item is „You buy „hot stocks‟ and avoid stocks that have performed poorly in the recent past.‟ Most of the investors assessed that this question was difficult to clearly understand The words
„hot stock‟ was vague and not suitable for recent stock markets in Vietnam To make it grasp more completely, they replaced „hot stocks‟ with „blue chip stocks‟, „midcap stocks‟ and „penny stocks‟ Therefore, the first item would be deleted and altered for three more questions: Question 1: You prefer buying „blue chip stocks‟ Question 2:
You prefer buying „midcap stocks‟ Question 3: „You prefer buying „penny stocks‟
Additionally, it should have question 4 that was separated from the first question:
„You avoid buying the stocks having poor performance in the recent past.‟ Moreover, it was impossible to analyze some representative stocks to make investment decisions for all stocks that they wanted to invest in They emphasized that the word „all‟ was not sufficient, and some other word was needed to replace on „the same as industrial field‟ Therefore the question „You use trend analysis of some representative stocks to make investment decisions for all stocks that you invest.‟ was changed to „You analyse effective operation of some representative companies to make investment decisions for all other stocks (the same as an industrial field)‟ Items of variable of Representative bias were presented as follows:
Table 3.1: Items of Representativeness REP Representative bias rep1 You prefer buying „blue chip stocks‟ rep2 You prefer buying „midcap stocks‟ rep3 You prefer buying „penny stocks‟ rep4 You avoid buying the stocks having poor performance in the recent past rep5 You analyse effective operation of some representative companies to make investment decisions for all other stocks (the same as an industrial field) rep6 You tend to choose stocks that are representative for VN30 in the stock market
3.3.2 Scales measurement of mental accounting bias
Variable of mental accounting bias is signed of MEN According to Waweru et al (2008), there are three items to measure this variable Most of investors agree with these questions They tended to treat each element of their investment portfolio separately because each stock had different features, type of business and industrial field Therefore, deciding to buy or sell, they often ignored the connection between different investment possibilities and evaluated a segregate factor and regardless of the aggregate investment portfolio Items of mental accounting bias were shown below:
Table 3.2: Items of mental accounting bias
MEN Mental accounting bias men1 You tend to treat each element of your investment portfolio separately men2 You ignore the connection between different investment possibilities men3 You evaluate a segregate factor and regardless of the aggregate investment portfolio
3.3.3 Scales measurement of gambler’s fallacy
Variable of gambler‟s fallacy is symbolized of GAM According to Waweru et al (2008), there are four items to measure this variable Most of the investors had the ability to predict reverse points to decide their investment because they had over 10- years of experience to invest stocks in the HOSE They also agreed that if the price of a stock decreased in many periods, they would decide to buy this stock because they believed that it would not go down any further Similarly, they supported the theory that if the price of a stock increased many times, they would sell this stock because they believed that it would not go up any higher They further explained that it was this rule of fluctuation and technical analysis „shoulder-head-shoulder‟ In the fourth question: „You determine to buy stocks regardless basing on fundamental or technical analysis.‟ Most of them disagreed to this question because it was impossible to invest in stocks without both fundamental and technical analysis The word „or‟ was not clear and easier to make the participants confused To survive today, investors had to have knowledge of fundamental and technical analysis Therefore this question was normal and less significant They suggested replacing another question: „You entirely focus on economic factors such as GDP, CPI, etc before deciding to invest stocks.‟
Almost they emphasized that it was popular to have a rumour of a stock that made investors pay attention to their investment decision because the current stock market was very sensitive The question hence was exposed: „You believe in rumours to decide to buy or sell stocks.‟ Particularly, this factor mentioned gambler‟s fallacy; it was based on belief and fortune Therefore, investors proposed two more questions :
„You decide to buy or sell stocks based on your forecast‟ and „You believe you are often lucky to invest in stocks‟ Items of gambler‟s fallacy were shown below:
Table 3.3: Items of gambler’s fallacy
GAM Gambler’s fallacy gam1 You guess reverse points to decide your investment gam2 Price of a stock has fallen in multiple sessions You determine to buy this stock because you believe that it will be impossible to decline more gam3 Price of a stock has increased in multiple sessions You determine to liquidate this stock because you believe that it will be impossible to go up more gam4 You much focus on economic factors such as GDP, CPI, etc before deciding to invest stocks gam5 You believe rumour to decide to buy or sell stocks gam6 You believe you are often lucky to invest stocks gam7 You decide to buy or sell stocks basing your forecast
3.3.4 Scales measurement of availability bias
Variable of availability bias is signed as AVAIL According to Waweru et al
(2008), there are three items to measure this variable As mentioned above, interviewees had over 10-years of experience in stock investment Therefore, they had a good knowledge of this field However, they were still afraid of investing in the foreign stock exchange and most of them prefered buying local stock compared to international stocks because the information of local stocks was more available
Moreover, they also recognized that the information from their close friends and relatives were as a reliable reference for their investment decisions because they said that they were sometimes busy, their close friends supplied valuable information to help them be more successful And it was certain that they were more confident to invest in stocks of companies where they had previous knowledge In short, items of availability bias did not change and were displayed below:
Table 3.4: Items of availability bias
AVAIL Availability bias avail1 You prefer to buy local stocks to international stocks because the information of local stocks is more available avail2 You consider the information from your close friends and relatives as the reliable reference for your investment decisions avail3 You tend to invest stocks of companies where you have worked or known
3.3.5 Scales measurement of herd behavior
Variable of herd behavior is signed as HERD According to Waweru et al
(2008), there are three items to measure this variable Most of interviewees listened to other investors‟ ideas and observe other investors‟ strategies In this stage, a sensitive stock market, they often considered other investors‟ decisions and were often affected by these decisions They were not only influenced by selecting types of stock and the number of stock shares, but also the buying and selling of stocks They often wondered why there were a lot of people who purchased this stock or sold that stock and whether there was unannounced good news or bad news regarding that stock that they had not known yet Therefore, it was better to base such decisions on the
“crowd” and they agreed that „unitied we stand, divided we fall.‟ In short, items of herd behavior maintain and were shown below:
Table 3.5: Items of herd behavior
HERD Herd behavior herd1 Other investors‟ decisions of choosing stock types have impact on your buying decisions herd2 Other investors‟ decisions of the stock volume have impact on your selling decisions herd3 X27 Other investors‟ decisions of buying and selling stocks have impact on your investment decisions
3.3.6 Scales measurement of over-underreaction
Data analysis approach
3.4.1 Test of scales measurement reliability
Cronbach’s Alpha Test is applied to test reliability of scales measurement with 5-point Likert measurements It also presents the consistency of a specific sample of participants‟ response through questions or items Additionally, it can assist to predict the reliability of respondents to the measurements (Helms, Henze, Sass &
Mifsud, 2006, p.633) Cronbach‟s alpha is usually applied for behavioral researches as an indicator of reliability (Liu, Wu & Zumbo, 2010, p.5) As a result, the Cronbach‟s alpha is completely suitable for this research because the questionnaire includes 5-point Likert measurements and the paper is related to behavioral finance
This paper employs Cronbach‟s alpha to test of scales measurement reliability including the factors that are built after the factor analysis Nunnally (1978) proposes that Cronbach‟s alpha should be at least 0.7 to make sure that the measurements are reliable (p.245) However, many statisticians believe that it can be acceptable if the Cronbach‟s alpha is over 0.6 (Shelby, 2011, p.143) Besides, statisticians also recommend that it is important to consider the corrected item-total correlations when using the Cronbach‟s alpha index The corrected item-total correlations, which reflect the correlation of items designated with the total score for all other items, should be at the acceptable score of 0.3 or higher (Shelby, 2011, p.143) This research selects the acceptable Cronbach‟s alpha is 0.7 – 0.8, with the corrected item-total correlation index is 0.3 or over 0.3 because the measurements of financial behavior are new to the stockholders of the Ho Chi Minh Stock Exchange Moreover, the accepted significant level of the F-test in Cronbach‟s alpha technique is not more than 0.05 The
Cronbach‟s alpha test is finished by SPSS software
In this study, EFA is used to explore the factors that the variables of behavioral finance of the questionnaire (question 12 to question 43) belong to EFA is used to reduce the number of items in the questionnaire that do not meet the criteria of the analysis (O‟brien, 2007, p.142) In this case, EFA is utilized to test the hypotheses shown in the research model of Chapter 3 In this research, the following criteria of the exploratory factor analysis are applied: Factor loadings, KMO, Total variance explained, and Eigenvalue Factor loadings are defined as correlations of each item with the factor that it belongs to Factor loadings of the items on a factor are greater than 0.5 (with the sample size is 100) ensure that EFA has a practical significance to the analyzed data (Hair et al., 1998, p.111) The Kaiser-Meyer Olkin Measure of Sampling Adequacy (KMO) presents the level of suitability of using EFA for the collected data The KMO should be between 0.5 and 1.0 (significant level less than
0.005) to make sure that factor analysis is suitable for the data (Ali, Zairi & Mahat,
2006, p.16) Total variance explained is used to identify the number of retained factors in which factors can be retained until the last factor represents a small proportion of the explained variance The total variance explained is suggested to be more than 50% (Hair et al., 1998, p.111) Eigen-value is an attribute of factors, being defined as the amount of variance in all items (variables) explained by a given factor
Eigen-value should be greater than 1 because Eigen-value is less than 1 means that information explained by the factor is less than by a single item (Leech, Barrett &
Morgan, 2005, p.82) The EFA is done by SPSS software.
DATA ANALYSIS AND FINDINGS
Data description
After collecting the data, the paper reported data background; It was initially concerned with gender: most investors were male, 75.46% with females at 24.54%
Secondly, it related to age: most investors were between 26 and 45 years old
Dissecting the data even more, 57.27% were between the age of 26 and 35 and 42.73% were between the ages of 36 and 45 Next was the work experience of the respondents: most investors had worked for less than 5 years with 10.09%, from 5 to
10 years accounting for 47.27%, and over 10 years work experience with 33.64%
The other background revealed was their average income Most investors (48.18%) earned from 12 to 20 million VND per month, 41.82 % earned from 6 to 12 million VND per month and approximately 10% earned over 20 million VND per month
Additionally, 35.45% of investors had invested funds in stocks for 5 to 10 years, while 39.09% from 3 to 5 years, and only 11.82% from 1 to 3 years Moreover, most investors ( 85.45%) had some type of formal training on how to invest money on the stock market Only 14.55% of the respondent investors did not take participate in any training courses about stock investment Finally, we asked about the total amount of money investors invested in HOSE last year There were 30.91% of investors invested from 60 to 120 million VND in stocks, 20.91% of investors joined in HOSE with a total amount from 120 to 300 million VND 14.55% of the investors spent from between 600 to 900 million VND and investing over 900 million was 11.82% of investors The summary of data background was shown in Appendix 5.4
Factor analysis of behavioral variables influencing the individual investment
Pilot survey aimed to test scales measurement reliability to define Cronbach‟s alpha of items from 0.7 to 0.8 to make sure that the measurements are reliable The questions of behavioral factors were from X12 to X38 that were independent variables Questions from Y39 to Y41 were created to identify the evaluation of investors about their own investment decision and they were dependent variables
Table 4.1: Idependent variables, dependent variables and items
1.Representativeness rep1, rep2, rep3, rep4, rep5, rep6 2.Mental accounting bias men1, men2, men3
3.Gambler‟s fallacy gam1, gam2, gam3, gam4, gam5, gam6 and gam7
4.Availability bias avail1, avail2 and avail3
5.Herding behavior herd1, herd2 and herd3
6.Over and under creation creat1, creat2 and creat3, creat4, creat5 and creat6
Investment decision return1, return2, return3
4.2.2 Results of Cronbach’s alpha analysis of pilot survey (NR)
The paper tested scales measurement reliability and its results For the independent variables, the first factor, representativeness had Cronbach‟s alpha of rep3 that was low ( 0.5 Therefore factor loading was suitable for data of survey Bartlett – Significant level is 0.000 less than 0.05
(0.0000.5), if different loadings between them more than 0.3 they would be accepted, whereas if different loadings between them less than 0.3, they would be rejected or deleted In the study, herd1, avail2 and creat1 did not meet this requirement ( 50%) so scales measurement was accepted In short, the author deleted five items including herd1, gam4, gam1, avail2 and creat Details of Cronbach‟s alpha analysis were presented in Appendix 5.5
Table 4.4: KMO and Bartlett’s Test, total variance explained and rotated component matrix (EFA time 1)
Initial Eigenvalues 1.511 (component 7) >1 => acceptable Rotation sums of squared loadings
Component 1 rep1, rep2, rep3, rep4, rep5 Component 2 herd3, herd2, creat5, herd1 herd1 deleted
Component 3 creat2, creat3, creat4 Component 4 return1, return2, return 3 Component 5 gam1,gam2, gam3, gam4, gam5 gam1 and gam4 deleted Component 6 avail1, avail2, avail3, creat1 creat1 and avail2 deleted Component 7 men1, men2, men3
After deleting items including herd1, gam1, gam4, avail2 and creat1, results presented that KMO = 0.544>0.5, Bartlett – Significant level was 0.000 (< 0.05);
Factor loadings were ≥ 0.5, all items were greater than 0.5 Additionally, the Rotate Factor matrix table, every item had some loadings more than 0.3, so these items were accepted Moreover total variance explained table showed how the variance was divided among the 22 possible factors Seven factors had initial eigenvalues =1.401 greater than 1, which was a common criterion for a factor to be useful Rotation sums of squared loadings = 71.362% > 50% so scales measurement was accepted Table 4.5 below presented details as follows:
Table 4.5: Summary of KMO and Bartlett’s Test, total variance explained and rotated component matrix (EFA time 2)
Initial Eigenvalues 1.401 (component 7) >1 => met requirement Rotation sums of squared loadings
The paper employed the SPSS software to divide into 7 components as Table 4.6 below showed All loadings of items cluster were greater than 0.5 and met requirement to group them into 7 components The paper finished exploration factor analysis and used these components for regression analysis Details of Cronbach‟s alpha analysis (time 2) were presented in Appendix 5.5
Table 4.6: Rotated component matrix (EFA time 2)
Component Items cluster Factor loadings
Regression analysis
According to Leech, Barrett and Morgan (2005), regression analysis is a statistical tool for the investigation of relationships between variables To explore such issues, the author collected data of variables and employed regression to estimate independent variables affect dependent variables Researchers used adjusted R 2 – lower than unadjusted R 2 – to indicate that how many percentages of the variance can be predicted from the independent And adjustment is affected by the magnitude of the effect and the sample size Besides that, the author also uses F -test to test significance of model Hypotheses of H0 and H1 are presented as follows:
H0: β1 = β2 = … = βk = 0 (no linear relationship) H1: at least one βi ≠ 0 (at least one independent variable affects Y)
If p-value
