U10_dq

Theoretical and Applied Economics Volume XXI (2014), No. 4(593), pp. 153-166

Behavioral finance: new research trends, socionomics and investor emotions

Adrian MITROI Bucharest University o f Economic Studies, Romania [email protected] Alexandru OPROIU Bucharest University o f Economic Studies, Romania [email protected]

Abstract. The p a p er presents a critique o f standard investment analysis, fundam ental and technical, ■ and develops an alternative more comprehensive approach that should include some o f the tenets o f behavioral finance. In the p ursuit o f understanding the behavior o f the m arket player, the basic argument relies on the supposition that the risk appetite increases exactly at the worst moment - when the capacity to assume additional risk decreases significantly. People view a sample randomly drawn fro m a population as highly representative and quasi sim ilar to the population in all its essential characteristics. They expect any two samples drawn fro m a particular population to be more sim ilar to one another and to the population than is statistically justifiable. This behavior is different fro m the tenets o f classic fin a n c e theory. The p a p er aims at demonstrating that investor psychological biases lead to investment perform ance to tilt to the mean in the long run and by fo llo w in g the trend, the fin a n cia l m arket population does not enjoy significant sustainable benefits. A s a reflection o f the behavioral biases and influences, the statistical demonstration supports the conclusion that markets do not random walk.

Keywords: psychology, biases, efficiency, individual investment.

JEL Classification: 5F, 5G, 5K, 7J, 7K,7L, 10B.10F. REL Classification :G 1 1 ,G 1 2 ,G 1 4 .

154 Adrian Mitroi, Alexandru Oproiu

Introduction

Behavioral finance does not eliminate but complements the standard evaluations approaches - fundamental, technical and markets analysis. It combines the findings of all valuation procedures with the investigation of social, psychological emotional aspects of the market, and relaxes the strict requirement of convergence between price and value. Since markets are always about high financial and social stakes, it is no wonder that most of the time the subjective emotions dominate the objective and logical approach.

Behavioral analysis considers the elements of human perception and evaluation of outside situation and events, and most importantly, the emotions associated, both ex ante and ex post with any financial decision. This new field of modem finance refers to neuroscience debate and assertion that the motivations, emotions, and feelings are indispensable to any human decision, including the financial ones; emotions are essential to any decision and course of action.

The investors interpret market data and events at two cognitive levels: the intellectual level of ordination, process and analysis of real factors (economic data), and the logical and rational level of understanding what this objective identifiable factors will influence the perception of the other players on the market. The information has investment value when is correlated with professional knowledge (human intellect) and interpersonal dynamics of market players (their emotions and sentiments). Due to uncertainty and continuous change in the game of the market, there is a strong interdependence between personal experiences (autobiographic memory) and rational expectations of the investors about the future, since their personal experiences influence the way they interpret and select available data.

Practically, behavioral finance complements but not replaces technical and fundamental analysis by the systematic analysis of the fundamentals of the market prices as a result of the correlation between investor experience and expectations and the market momentum. Slow changes in the market sentiment are not emotionally contagious, but they insinuate slowly in a market trend. Sudden moves, on the other side, are attributed to new strong evidence presented and disseminated by the market; these do not have a lasting effect, once the new resistance or support floor was established. In general, investors tend to accept with relative ease the market momentum that is imposed by the majority rather than adopt a contrarian investment strategy, since dissatisfaction of a negative result, after a contrarian decision, weighs significantly higher than the eventual satisfaction coming from a contrarian decision. Practically, the rational reasoning guides investor behavior based on the network of commercial interests and economic compromises (as unintentional consequences price formation auctions

Behavioral finance: new research trends, socionomics and investor emotions 155

o f the m arket participants) and m aintenance o f a stable em otional status quo, lack o f internal and external contradictions (em otional dissonance) that can sanction the intellect o f the investor to form and im plem ent an investm ent tactic that is feasible, easy to understand, am end and m anage w ithin the risk boundaries that are generally accepted.

Database fo r the paper

The sources o f this data include Stock M arket C onfidence Indexes - a s linked from Y ale School o f M anagem ent International C enter for Finance as directed by the 2013 N obel Prize in Econom ics Dr. Robert Shiller - the Investor B ehavior Project. A dditionally, S& P/C ase-Shiller H om e Price Indices is a key source o f data.

A nother, significant input o f data is relatively easily dow nloadable from Shiller, R., U .S. Stock Price D ata, A nnual, w ith consum ption, b oth short and long rates, and present value calculations, (h ttp://aida.w ss.yale.edu/~ shiller/data.htm )

Nobel's inter-disciplinary connections T he 2013 N obel has forced us further in the inter-disciplinary path w here fundam entalists, statisticians, psychologists, physicists m ust w o rk hand in hand to find new universal laws w hich can select. Som ehow m y inter-disciplinary m ind registered Eugene faster than Fam a. A fter all, Eugene Stanley, the father o f econo-physics, could also g et a N obel. I f psychologists could g et the biggest aw ard for econom ics, a p hysicist could have been there, too. B ut then the surprise becam e bigger, n ot because it w as Fam a n ot Stanley, b ut because R o b ert Shiller shared the award. W hen behavioral finance got the N obel for econom ics in 2005, the E conom ist m agazine carried an article pointing out how a new theory had ju n k e d 200-year old classical econom ics.

It w as n ot ju s t m edia, bu t even psychologists w ho w ere b ent on b u rying classical econom ics. E fficient m arket hypothesis w as presum ed dead. It w as considered deficient. B ut over the years, defending the underdog changed to understanding the n ew theories and then finally even questioning them (S hifter’s exuberance, end o f behavioral finance). I t’s a fight betw een perception and reality at a certain p o in t o f tim e, w hich o f course is dynam ic, leading to new perceptions and new realities at new points in time. N o w that F am a has b een acknow ledged y et again, his tough stand against behavioral finance as stories o f anom alies can be seen in m ild er light. A fter aft, standing there w ith Shifter w ould definitely m ake him believe “ even to g eth er w e d o n ’t know aft the tru th ” . T he blind m en and the elephant m etaphor rem ains a strong them e. ‘M y elephant is efficient w hile yours is in efficien t’ has been overruled by the N obel com m ittee w hich believes that the elephant is both efficient and inefficient som etim es.

156 Adrian Mitroi, Alexandru Oproiu

On one side Shiller’s exuberance is quite clear from the illustration above, how fundamental earnings diverge from real prices. But on the other side a two scale look (illustration below) shows some similarity in growth and decay seasonality among the two values. The seasonality aspect is not discussed in Shiller’s work The failure of behavioral finance to take it from the fundamentalists can be viewed as a victory of sorts, but it’s still an illusion. The Nobel Prize bashers like Taleb also won’t enjoy this as their randomness theories get weaker by the day. The committee needed Lars Peter Hansen to balance. What did Hansen do? Hansen says that both efficient and inefficient schools could be understood better with more testing as the economic system is not static, it’s a dynamic system with multiple moments.

Is this not a step ahead toward assuming markets as natural systems? Are the laureates not struggling to understand divergence: why are markets not predictable in the short term and why are they predictable in the long term?

Year

Source: Robert Shiller's plot of the S&P Composite Real Price Index, Earnings, Dividends, and interest Rates, from Irrational Exuberance, 2nd edition, 2005.

Behavioral finance: new research trends, socionomics and investor emotions 157

Date S&P 500 PE S&P 500 Price Real 1/1/2013 21.9 1600 1/1/2000 43.77 800 1/1/1390 17.05 600 1/1/1980 8.85 200 1/1/1970 17.09 600 1/1/1960 18.34 400 1/1/1950 10.75 200 1/1/1940 16.37 100 1/1/1930 22.3 400 1/1/1920 5.99 30 1/1/1900 18.63 200

Source: S&P Composite Real Price-Earnings Ratio and Interest Rates using data from irrationalexuberance.com/shiller downloads/ie data.xl

Today’s volatile markets systematically complicated the decision making process, especially on the money management dilemmas. The stress hormones, adrenaline and cortisone, although highly efficient defense mechanism from an evolutionary perspective, can severely impair our ability to make clear-head decisions in the daily fragility and volatility o f the stock market.

When we, as investors, are nervous and stressed, our ability to think clear diminishes and we become more pessimistic, lose our ability to think clearly and concisely, and become more impatient. Increased level o f stress, in line with financial, professional and personal high stakes involved in our investment decisions, lead to biased information-processing mistakes, like overtrading, overconfidence or illusion o f control. Our behavior biases and tendencies can harm the investment performance o f our portfolio. We feel contented by the illusion o f action. Subjected to stress, our brain would rather favor action than inaction, thinking and planning. Rational analysis mixes with up with our gut reaction and prejudices and we often forget that there are no short cuts to the places really worth going. Misses Market has no indulgence on our urge to make quick money.

In the article „Emotional Intelligence and Investor Behavior” o f John Ameriks, Tanja Wranik and Peter Salovey conducted a survey on private held portfolios. Investment performance correlates positively with high Emotional Intelligence (recognition and usage o f one’s emotions in a productive manner) scores. However, the main presupposition o f intelligent investing is that a constant supply o f counterparties trade in the market, probably less emotionally and financial intelligent. Common sense would relate investor emotional intelligence with a better ability for decision making, and consequently a more systematic and disciplined approach on managing financial affairs. A person with high emotional intelligence, is by definition, an individual that is able to identify, comprehend, and regulates her emotions in decision making and problem solving. Personality and emotional features o f an individual have both destructive and destructive effects in

158 Adrian Mitroi, Alexandru Oproiu

the financial decision making process. Investment decisions are often constrained by time pressure, social rules and regulations, and the continuous change and uncertainty o f the market place. As such, monetary and financial decisions are significantly influenced by psychological factors.

The topics presented make reference to ideas and research o f the great savants o f modem finance. They range from Kahneman and Tversky (asymmetric predisposition o f individuals response to losses and gains), Sheffin, Statman, Barber, Odean (selling winning investments too early and keeping the losing ones too much), Statman (theory o f regret), Gilovich, Griffin, Kahneman (cognitive heuristics and biases) and others. One critical assumption is to consider the return distribution as normal or lognormal. In a normal distribution, return distribution intervals have a constant measure, in a lognormal distribution, the intervals value depends on the relative value o f stock price. I f stock price variations are independent, the return distribution is normal, and if log differences are independent and have a finite variance, the price distribution is lognormal. An efficient, fair market should preclude any investor to infer immediate evolution based on past evolution (Bodie et al., p. 342).

Robert Shiller’s ’ Paper on ‘The Volatility o f Stock markets P rices’ published in 1987 uses dividend data and real interest rates to seek evidence that true investment value changes through time sufficiently to justify the price changes. His paper concluded that most o f the volatility o f the stock market prices appears unexplained.

Shiller volatility or fluctuations prove that behavior o f markets is not normal. Non normal distribution series is a widely followed proof o f inefficiency in prices:

3 3Ol■*->

t / 3

U t <D e

0)H i

b f)

3

Behavioral finance: new research trends, socionomics and investor emotions 159

The mean reversion theory suggests that prices and returns eventually move back towards the mean or average. This mean or average can be the historical average of the price or return or another relevant average such as the growth in the economy or the average return of an industry.

This theory has led to many investing strategies involving the purchase or sale of stocks or other securities whose recent performance has greatly differed from their historical averages. However, a change in returns could be a sign that the company no longer has the same prospects it once did, in which case it is less likely that mean reversion will occur. Percent returns and prices are not the only measures seen as mean reverting; interest rates or even the price-earnings ratio of a company can be subject to this phenomenon.

tAV"'

However many times after falling below mean markets don’t revert back to the absolute mean (illustration b).

While sometimes after getting oversold (overvalued) and staying above the mean attempting to get back to mean value, markets or asset prices stay overbought and get more overbought (overvalued) (illustration c).

Sometimes there is a clear disregard to mean value, markets ignore the mean totally (illustration d).

A continuous mean-reverting time series can be represented by an Omstein- Uhlenbeck stochastic differential equation dxt = 0(p - xt)dt + odWt

Figure 1. How markets pressure to mean reversion

Ideal mean reversion in illustration ’a’ is how markets should express mean reversion. The overbought (overvalued should push back to an absolute mean and vice versa). Where 0 is the rate of reversion to the mean, p is the mean value of the process, o is the variance of the process and Wt is a Wiener Process or Brownian Motion. In a discrete setting the equation states that the change of the price series in the next time period is proportional to the difference between the mean price and the current price, with the addition of Gaussian noise. One critical assumption is to consider the return distribution as normal or lognormal. In a

160 Adrian Mitroi, Alexandru Oproiu

normal distribution, return distribution intervals have a constant measure, in a lognormal distribution, the intervals value depends on the relative value of stock price. If stock price variations are independent, the return distribution is normal, and if log differences are independent and have a finite variance, the price distribution is lognonnal.

An efficient, fair market should preclude any investor to infer immediate evolution based on past evolution (Bodie et al., p. 342). The local market research introduced also by the paper tests the random walk hypothesis to see if markets move at random and investors do not express any behavioral biases. To test daily return distribution and independence, a regression equation is introduced: \n{It) = jj + p \n (It_x) + £t

Where: I, is the index value (the most representative, BET Index) in day t and st is the residual. Next, to test the linear dependence, paper introduces the regression: e, = <f>0 + >̂l£ t_l

If <j\ proves to have statistical significance than we can conclude with the degree of confidence that the evolution is linear dependent, the market does not follow a random walk. Then, the nonlinear dependence is tested by GARCH models ARCH general model (GARCH (p,q)): r, = Pq+ PUrt) + £t £t *N( 0, ht ) h, = a Q+a(L)£* +y(L)h,

Where: r, is an ARMA process (p’,q’)(1) (or AR(p’) or MA(q’)); h, is an ARCH(p) process and GARCH(q).

The anticipatory behavior o f most investors, who would rather take into account the market developments rather than the financial and economic performance of a company, is predominant. We were excited by the idea, first introduced by the paper of Stancu and Stancu (2013), “Rationality versus Irrationality on the Romanian Capital Market”, where the authors contend that: ‘The shares o f financial services companies confirm the second case, that o f

irrational, subjective behavior, not only at the level o f the individual investor

B e h a vio ra l finance: new research trends, s o cio n om ics and in v e s to r em o tio n s 161

but also at the level o f the community o f stock exchange operators. The shares are traded mostly fo r short-term gains purpose. Their stock prices reflect investor expectations o f the stock market development and not the issuer's financial performance. As a consequence, investors have the priori belief that these performances are greater that they are in reality. With that in mind, their concern is purely speculative ”. We have chosen to test BET and BETC as an interesting indicator of the investor over confidence in their prediction power.

The author’s conclusion on real economy issuers vs. nominal economy companies: “The share o f successful manufacturing companies confirms the rational economic behavior, consistent with fundamental financial analysis. This conclusion is reinforced by the fa ct that these companies have evidently stable numbers o f trade and significant volumes o f sales on the stock market. Moreover, the financial results are obtained based on tangible products sales, which gives confidence in their stability. This is the result o f long-term investments in their assets. Other companies in the manufacturing industry, characterized by large fluctuations in turnover and lower trading volumes on the stock exchange, reject the assumption o f rational economic behavior. The investors sanction the instability o f these companies financial activity and their volatility o f the stock price volatility”. This academic path as interesting and a prime on behavioral finance literature: Not only personal and individual circumstances or market and contextual influence converge to an investment decision; a third factor can modify the investor perception, i.e., what type of company is the focus of investment analysis. Different sectors have different life cycles, but most important, sectors swap places in investor’s scope from great interest (like) to complete disregard (dislike). These new dimensions add an informative angle to investment and behavioral portfolio management. For the locally available index BET(2).

With the following specifications: rt is AR (1), ht is ARCH (1), GARCH (1) cu asymmetry factor for lag 1, the result is TGARCH (1,1,1):

162 Adrian Mitroi, Alexandru Oproiu

Dependent Variable: D_L_BET Method: ML - ARCH (Marquardt) - Generalized error distribution (GED) Sample (adjusted): 3 2721 Included observations: 2719 after adjustments Convergence achieved after 17 iterations Variance backcast: ON GARCH = C(3) + C(4)*RESID(-1)A2 + C(5)*RESID(-1 )A2*(RESID(-1 )<0)

+ C(6)*GARCH(-1)

Coefficient Std. Error z-Statistic Prob.

C 0.000481 0.000191 2.520458 0.0117 D_L_BET(-1) 0.194777 0.019558 9.958822 0.0000

Variance Equation

C 2.69E-05 3.67E-06 7.312937 0.0000 RESID(-1 )A2 0.316145 0.039609 7.981630 0.0000

RESID(-1 )A2*(RESID(-1 )<0) 0.092534 0.057950 1.596802 0.1103 GARCH(-1) 0.557362 0.036212 15.39155 0.0000

GED PARAMETER 1.193076 0.033061 36.08695 0.0000

R-squared 0.027324 Mean dependent var 0.000142 Adjusted R-squared 0.025172 S.D. dependent var 0.015930 S.E. of regression 0.015728 Akaike info criterion -5.908766 Sum squared resid 0.670881 Schwarz criterion -5.893556 Log likelihood 8039.967 F-statistic 12.69742 Durbin-Watson stat 2.061150 Prob(F-statistic) 0.000000

This regression is statistically significant, so we can conclude that a significant nonlinear dependence exists between daily returns on that specific index, BET. We can confidently assume that index pattern evolution does not follow a random walk. The local market research introduced also by the paper tests the random walk hypothesis to see if markets move at random and investors do not express any behavioral biases. We can infer that there is linear dependence between daily returns, and the index series of BET does not follow a random walk pattern. Other factors could influence this evolution, and they are persistent and consistent. For a second index, more comprehensive/composite, BETC:

Behavioral finance: new research trends, socionomics and investor emotions 163

BETC 3)

Dependent Variable: DJ.J3ETC Method: ML - ARCH (Marquardt) - Generalized error distribution (GED) Sample (adjusted): 3 2721, Included observations: 2719 after adjustments Convergence achieved after 14 iterations Variance backcast: ON GED parameter fixed at 1.5

GARCH = C(3) + C(4)*RESID(-1)A2 + C(5)*RESID(-1 )A2*(RESID(-1 )<0) + C(6)*GARCH(-1)

Coefficient Std. Error z-Statistic Prob.

c 0.000521 0.000217 2.400159 0.0164 D_L_BETC(-1) 0.200854 0.020807 9.653185 0.0000

Variance Equation

C RESID(-1)A2

RESID(-1 )A2*(RESID(-1 )<0) GARCH(-1)

2.86E-05 0.308604 0.090202 0.537597

2.67E-06 0.028024 0.043393 0.027784

10.68003 11.01231 2.078745 19.34889

0.0000 0.0000 0.0376 0.0000

R-squared 0.026846 Mean dependent var 0.000142 Adjusted R-squared 0.025053 S.D.dependent var 0.015930 S.E. of regression 0.015729 Akaike info criterion -5.891979 Sum squared resid 0.671210 Schwarz criterion -5.878942 Log likelihood 8016.145 F-statistic 14.96863 Durbin-Watson stat 2.073197 Prob(F-statistic) 0.000000

For this index also we can construe that this regression is statistically significant, so we can conclude that a significant nonlinear dependence exists between daily returns on that specific index, BETC. We can confidently assume that index pattern evolution does not follow a random walk. We can infer that there is linear dependence between daily returns, and the index series of BETC does not follow a random walk pattern. Other factors could influence this evolution, and they are persistent and consistent. For this nonlinear dependence, the model specifications were: rt is AR (1), ht is ARCH (1), GARCH (1) with asymmetry factor for lag 1, the result is 1, 1, 1). We can confidently assume that BETC index pattern evolution does not follow a random walk.

C o n c lu s io n s

Research in behavioral finance has important practical and academic applications. The research can help guide investment portfolio allocation decisions, both by helping the understanding the kinds of errors that investors tend to make in managing their portfolios, and also by allowing us to understand better how to allocate assets and locate profit opportunities for investment managers.

164 Adrian Mitroi, Alexandru Oproiu

Understanding the psychological foundation of human behavior in financial markets facilitates the formulation of investment policy statements for individual investors. Methods that originate in psychology are used as research tools, along with traditional finance research methods. Over these years, the academic and practioners world of finance have seen the blossoming of behavioral finance into a significant body of knowledge. The combination of theoretical and empirical work has allowed us to see the relevance of the basic psychological theories to many financial phenomena. The newly developed body of knowledge is an important addition to the theory and practice of modem finance. People tend to discount the eventual implications of low probability- high negative impact events, but these events, due to their apparent low probability, seem to happen less often than anticipated. The most expected outcome of these possible yet less probable events can have, however, disastrous effects on the prospect value of investor portfolio. High emotional impact events, although rare, have a major, indelible impact on the emotional registry of a person. Any subsequent decision is affected by historical record of successes and failures. In general, investor that succeeds and survives on the long term, makes small gains systematically (or wins more and more times than they lose). Their investment success is not a simple luck of result of a continuous stream of rational and correct material decisions but of a disciplined and focused approach, prime access to information and ability to assemble on time and correctly the available data, coupled with the ability for innovation and adaptation to the continuous change and challenge of the market game. Investment managers have to prove their repeatable professional ability and sustainable value-adding capability on a continuous basis to their employers, employees, and investment public. Although the business of managing investment assets is much more complicated, competitive, rewarding and challenging than ever, and investors are increasingly sophisticated, their emotional attributes remain as simple as always - fear of losses and desire to make money. Mental

•cognitive errors are frequently caused by heuristic simplifications - logical shortcuts by which decision makers use simple rules to solve complex problems. When this approach is used inappropriately for complex problems solving, investors’ biases could lead to systematic mental mistakes. The paper advances the idea that the investor psychological biases lead to investment performance to tilt to the mean in the long run and by following the trend, the financial market population do not enjoy significant sustainable benefits. As a reflection of the behavioral biases and influences, the statistical demonstration supports the conclusion that markets do not random walk.

The article reviews some psychological concepts relevant and used in the study, in an interdisciplinary effort of understanding the correlation or causality between

Behavioral finance: new research trends, socionomics and investor emotions 165

psychology and finance. The paper aims at demonstrating whether investor psychological biases lead to investment performance to tilt to the mean in the long run. As a reflection of the behavioral biases and influences, the statistical demonstration supports the conclusion that markets do not random walk. By following the trend, the financial market populations do not enjoy significant sustainable benefits. In the research reported here investigated the market pattern zigzag to see any predilections or biases or a random walk. Analyzing the data for this study leads to the interesting conclusion that individual psychological biases and differences should not be confounded with noise within econometric models but rather manifest a solid influential role on the dependent variable - investment outcome. Data base source for the article shows that psychological characteristics have salient relationships with various aspects of investment decision making process making and the transactional activity of the individual investor.

The statistical interrogation describes the sampling methodology, the frequency of data and the empirical methodology that lead to analysis of the results and concluding remarks. The study provides details on raw statistical test scores, regression results and analysis. In this study, I evaluate the association between investors’ behavior and her portfolio results. The findings suggest that psychological biases can have an impact on risk return optimization, asset allocation on investment portfolios and finally on investment outcome. The sources of investor biases that lead to investor finance errors the investment management industry can apply the data for the development of products and services (automated pilot investing) that may help save investors from sabotaging their financial standing and future prospects.

Also, new behavioral portfolio construction methods should combine evidently classic finance math with rigorously quantified psychological metrics to improve models for operators use in giving financial advice and crate investor portfolios that enhance investors chances for reaching their life time financial goals. The future research

Students of Behavioral Finance still have much to research on influence of psychological profile dissimilarities between individuals and how these dissimilarities manifest in real financial investment decision and behavior. Personality and other individual circumstances and differences systematically influence investment decisions.

166 Adrian Mitroi, Alexandru Oproiu

Notes________

(1) General format of an AR process witli a finite number of p variables contribute to the current level o f y variable: y t = axy t_x ■¥a2y t_1 + .:+ a py,_p + S, , where a, are the coefficients to be estimated and S

represents the random residual in a classical regression equation, s , = y t (l — a f 1 + a 2L2 + ... + a pLp) .

L1 is a lag operator that for the value of the variable for the period and for the current period. The MA q

process of a q ranking can be arranged as follows: y t = e t - ^ b ; S t_j and also can be expressed based i=l

on time lag: ARMAprocess: y t (1 — b,L* + b 2L2 + ... + b qLq) = £t (l - b[L' + b 2L2 + ... + b qLq) . (2) BET index public data: Apr. 1998 - Ian. 2012. (3) BET index public data: A pr. 1998 - Ian. 2012.

References

Bodie, Z., Kane, A., M arcus, A.J. (1989). Investments, M cGraw-Hill Fam a, E.F., Fischer, L., Jensen, C. and Roll R. (1969). ’’The adjustm ent o f stock prices to new

info rm atio n ”. International Economic Review, vol. 10/1 K ahnem an, D. (2003). “M aps o f bounded rationality: A perspective on intuitive jud g m en t and

choice”, [Nobel Prizes 2002], Stockholm, Sweden M alkiel, B.G. (1973). A Random Walk Down Wall Street, N orton & Co M arkow itz, Harry. (1952). “Portfolio Selection: Efficient Diversification o f Investm ents”, Journal

o f Finance, Cowles Foundation M onograph: No. 16 M itroi, A., Stancu, I., (2007). “Behavioral finance vs. fundam ental and technical analysis”,

Symposion Scientifique International, Developpement durable, echange global et ecosystems, project AUF 6 3 1P43

Pal, M ., M itroi, A., Shah, M. (2011). “Tem poral Changes in Shiller's Exuberance D ata”, in Social Science Research Network, http://ssm .com 1754388, h ttp ://ssm .com /abstract=l754388 Econometrics: Data Collection and Data Estimation Methodology’ eJournal

Shefrin, Hersh and Statman, Meir. (2000). “ Behavioral Portfolio Theory” . The Journal o f Financial and Quantitative Analysis, Vol. 35, No. 2 (Jun.)

Shiller, R.J. (2001). “B ubbles, Hum an judgm ent and expert opinion”. CFA Institute Financial Analysts Journal, June 2002, Vol. 58, No. 3, 18-26

Stancu, I., Stancu, D., (2013). “Rationality versus irrationality on the Romanian capital m arket”, Theoretical and Applied Economics, Special Issue, available at: http://store.ectap.ro/suplim ente /Intem ational_F inance_and_Banking_Conference_FI_B A _ 2 0 13_X Ith_Ed.pdf

Statman, M. (2003). “B ubbles and portfolios: Investm ent lessons from Kahnem an and Smith, w inners o f the 2002 N obel Prize in Econom ics” Santa C lara University

Copyright of Theoretical & Applied Economics is the property of General Association of Economists in Romania and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.