Social Icons

twitterfacebookgoogle pluslinkedinrss feedemail

Sunday, 10 February 2013


Behaviors and opinions of single individuals may be influenced by the dynamics operating within the groups where they interact. Likely, this influence is stronger when uncertainty characterizes the decisional stage [1].  The conformity to common behaviors  has been explained originally in ethology by the selfish herd theory as an attempt to reduce the predation risk [2]. In human environment, the herd behaviors may take place in several situations and may affect deeply large groups of population. This is the case, for example, of the so called boom and bust phases in the stock markets. An explanation of the molding of a common decision (or opinion) within the financial markets has been given by the thought contagion theories, whereby single investors decide their operations by following the trend of the market. Thought contagion theories arose from the studies on epidemiological diffusion of pathologies within a population [3] and then they have been adapted to analyze the transmission of beliefs and information flows in the stock markets [4-6]. Propagation of the thought contagion would rely on three factors [4]: 1) transmission rate of beliefs/opinions; 2) receptiveness and 3) duration. The transmission rate measures how freely and frequently the subjects share their own belief/opinion or act coherently to their belief/opinion so as their behaviors become an information. Receptiveness indicates how likely subjects who originally do not have a certain belief/opinion, are then disposed to follow it, that is, the openness to new ideas. Duration denotes how long those who have a certain belief/opinion keep on transmitting this information to the others. The thought contagion speeds up the spreading of decision/opinion among the members of that group. In some sense, the herd behavior by leading to a common decision/opinion in a group contributes to improve the efficiency of the information flow. Both under ideal conditions and under conditions of uncertainty, the conformity to the common behavior may result as rational. In the ideal condition, where no one retains privileged information and all the possible information is made available to the whole group (in the context of financial markets this condition corresponds to the realization of the strong hypothesis of market efficiency as defined in the theory of efficient market [7]), following the information cascade is a rational strategy because given that everybody shares the same information, then it is unlikely for a single to outperform the common decision (e.g., the market trend). 
In sequential DM, where the subjects are informed about the final decisions taken by their predecessors, the subjects tend to imitate the former choices by elaborating the probability of success of the flow of available information. This leads to a reasonable convergence of preferences and to herding behavior [8-10]. The conformity to this common preference takes place if an information cascade has been recognized. That is, when the subjects, after having observed the actions of their predecessors, converge to the same choice that the others have made, independently of their own private information signals. Following a cascade means that people discard their own information in favor of deductions based on earlier people’s decisions [11]. The analysis of the capability of the subjects to exploit the whole information set in order to take their decision upon a relatively complex problem emerges as a relevant topic within the decision theory.
Under condition of uncertainty, i.e., in conditions of limited information, provided that the imitation of the others’ behavior is not mindless but relies on the bayesian approach, the herd behavior arises from rational inferences [12]. Predictions about the event are made sequentially and the very first decisions in the chain tend to reveal the private signals that are highly informative. In this sense the conformity to the initial pattern of decisions (even though in conflict with the hint that the private signal gives) makes the  behavior of the followers as rational [13-15]

Herding behavior: under certain conditions, it is rational for an individual to follow the crowd even if the individual’s own information suggests an alternative choice

Since to adhere to the information cascade originates from Bayesian induction, this strategy is considered to be rational. Deviations from the  information cascade gives rise to the so-called overconfidence effect, for which under certain conditions, the decisions in a sequential decision making would be otherwise determined by private information. 

On the contrary, this mechanism for the transmission of beliefs and opinions is expected to introduce some biases or even irrationality (e.g., in the fixing of stock prices) when not all the three conditions above mentioned are fulfilled. This may occur if: a) one or few subjects still retain privileged information by the effect of time lags or breakdown in the transmission; b) information is incomplete, unfounded or fake; c) there exists receptivity viscosity.
On the characteristics of the latter has been focused this work. Deviations from the information cascade are classified as overconfident. Therefore, the overconfidence effect occurs when subjects do not switch from their private information signals to the common decision because they value that their own judgment outweighs the others’ one
It is usually reported in literature the paradigm of inverse relation between overconfidence and accuracy: to the high confidence to one’s own judgement would not correspond high accuracy of the results. That is, self-assessment of accuracy in the cognitive domain produces overconfidence [15]. This inaccurate judgment (that may be also associated to perceptual distortion or illogical interpretations of facts/data) then would give rise to a cognitive bias which is called irrationality [16,17]. As consequence,  the overconfidence has been associated to irrational decisional process and to negative implications (also in a wide scale), which make the study on this effect a relevant topic in neuroeconomics. 

However, some experimental results suggest different keys to interpretation of overconfidence. In fact, performances in the overconfident group were found similar to the ones obtained from analytical (i.e., rational) tasks. Maybe the deviations from the cascade reveal that under some circumstances subjects have difficulty with recognizing the additional information provided by the public sequential decisions. 

  1. Ghosh D., Ray M. (1997). Risky ambiguity and decision choice: Some additional evidence. Decision Sciences, 28, 81-103.
  2. Hamilton, W. D. (1970). Geometry for the Selfish Herd. Diss. Imperial College, London.
  3. Bayley N.T. (1957). The mathematical theory of epidemics. London: C. Griffin.
  4. Lynch A (1998). Units, events, and dynamics in memetic evolution. Journal of Memetics - Evolutionary Models of Information Transmission, 2, 61-69.
  5. Lynch A. (2000). Thought contagions in the stock market. The Journal of Psychology and Financial Markets, 1, 10-23.
  6. Shiller R.J. (2000). Irrational Exuberance. Princeton: Princeton University Press.
  7. Fama, E.F. (1965). The Behavior of Stock Market Prices. Journal of Business 38(1): 34-105.
  8. Smith V.L., Suchanek G.L., Williams, A.W. (1988). Bubbles, crashes, and endogenous expectations in experimental spot asset markets. Econometrica, 56, 1119-1151.
  9. De Bondt W.F., Forbes W. (1999). Herding in analyst earnings forecasts: Evidence from the United Kingdom. European Financial Management, 5, 143-163.
  10. Prechter R.R. (2001). Unconscious herding behavior as the psychological basis of financial market 
  11. Easley, D., Kleinberg, J. (2010). Networks, Crowds, and Markets: Reasoning about a Highly Connected World. Cambridge University Press.
  12. Banerjee, A.V. (1992). A simple model of herd behavior. Q J Econ 107, 797–817.
  13. Bikhachandani, S., Hirshleifer, D., Welch, I. (1992). A theory of fads, fashion, custom and cultural change as informational cascades. J Polit Econ 100, 992-1026.  
  14. Anderson, L.R., Holt, C.A. (1997). Information cascades in the laboratory. Am Econ Rev 87(5), 847-862.
  15. Pallier, G., Gerry, R., Danthiir, V., Klietman, S. (2002). The role of individual differences in the accuracy of confidence judgments. J Gen Psychol 129(3): 257-299.
  16. Kahneman, D., Tversky, A. (1972). Subjective probability: A judgment of representativeness". Cogn Psychol. 3 (3): 430–454.
  17. Ariely, D. (2008). Predictably irrational: the hidden forces that shape our decisions. Harper Collins, New York.

No comments:

Post a Comment