OVERCONFIDENCE PHENOMENON

 

"No problem in judgment and decision making is more prevalent and more potentially catastrophic than overconfidence". Plous (1993)

"People are overconfident. Psychologists have determined that overconfidence causes people to overestimate their knowledge, underestimate risks, and exaggerate their ability to control events. Does overconfidence occur in investment decision making? Security selection is a difficult task. It is precisely this type of task at which people exhibit the greatest overconfidence." Nofsinger (2001)

·         "Overconfidence is greatest when accuracy is near chance levels.

·         Overconfidence diminishes as accuracy increases from 50 to 80 percent, and once accuracy exceeds 80 percent, people often become under confident. In other words, the gap between accuracy and confidence is smallest when accuracy is around 80 percent, and it grows larger as accuracy departs from this level.

·         Discrepancies between accuracy and confidence are not related to a decision maker's intelligence." Plous (1993)

"Overconfidence and anchoring definitely appear to be part of the explanation underlying post-earnings-announcement drift." Shefrin (2000)

"There are two main implications of investor overconfidence. The first is that investors take bad bets because they fail to realize that they are at an informational disadvantage. The second is that they trade more frequently than is prudent, which leads to excessive trading volume." Shefrin (2000)

"The classic study in over-confidence is Lichenstein, Fischoff and Philips (1977)."
Montier (2002)

"Overconfidence, however generated, appears to be a fundamental factor promoting the high volume of trade we observe in speculative markets. Without such overconfidence, one would think that there would be little trading in financial markets." Shiller (2000)

THE INTELLECTUAL CONCEIT EVIDENT IN OUR JUDGMENTS

The intellectual conceit evident in our judgments of our past knowledge (“I knew it all along”) extends to estimates of our current knowledge. Daniel Kahneman and Amos Tversky (1979) gave people factual questions and aksed them to fill in the blanks, as in: “I feel 98 percent certain that the air distance between New Delhi and Beijing is more tan ____ miles but less than ____ miles.”

Most subjects were overconfident: About 30 percent of the time, the correct answers lay outside the range they felt 98 percent confident about. Baruch Fischhoff and his colleagues (1977) discovered the same overconfidence phenomenon when people rate their certainty about their answers to multiple-choice questions, such as: “Which is longer (a0 the Panama Canal, or (b) the Suez Canal?” If people 60 percent of the time answer such a question correctly they will typically feel about 75 percent sure.

To find out whether overconfidence extends to social judgments, David Dunning and his associates (1990) created a little game show. They asked Stanford University students to guess a stranger’s answers to a series of questions, such as, “Would you prepare for a difficult exam alone or with others?” and “Would you rate your lecture notes as neat or messy?” Knowing the type questions, but not the actual questions, the subjects first interviewed their target person about background, hobbies, academic interests, aspirations, astrological sign anything they thought might be helpful. Then while the target persons answered 20 of the two-choice questions, the interviewers predicted their target’s answers and rated their own confidence.

SIX ERRORS AND BIASES IN HUMAN THINKING AND PROBLEM SOLVING

The confirmation bias:                    This is the tendency to search primarily for information that confirms our preconceptions. We discussed this phenomenon in class. Suppose you give subjects a sequence of three numbers {2, 4, 6} and ask them to come up with the rule for the sequence (the actual rule is any three ascending numbers). To discover the rule, they are allowed to generate sequences which are then either confirmed our disconfirmed. This is done until the subject thinks the rule has been found. Most often, people initially hypothesize that the rule is consecutive even numbers. As a result, they generate sample sequences such as {8, 10, 12} which confirm the hypothesis. They do not generate sequences which might disprove their hypothesis. Thus, most of the time the subjects guess the wrong rule.

The representativeness heuristic:  This is a thinking strategy people use to judge the liklihood of something by how well it matches a particular prototype. It leads people to ignore useful information. Here's an example from Myers (p. 53): Consider Linda, who is 31, single, outspoken, and very bright. She majored in philosophy in college. As a student she was deeply concerned with discrimination and other social issues, and she participated in antinuclear demonstrations. Based on this description, which would you say is more likely: (a) Linda is a bank teller or (b) Linda is a bank teller and active in the feminist movement. Most people answer (b) because it matches their impression of Linda. But this is an error because it is not possible for the conjunction of two events to be more likely than one of the events alone. So (b) can never be more likely than (a).

Using usless information:                Here's another example from Myers (p. 53) . Consider the following two questions: Roberta is a university student who spends about three hours studying outside of classes in an average week. What would you guess her grade point average to be? Judith is a university student who spends about three hours studying outside of classed in an average week. Judith has four plants in the place she's living now. On an average weekday, she goes to sleep around midnight. She has a brother and two sisters. Two months is the longest period of time she has dated one person. She describes herself as being often a cheerful person. What would you guess her grade point average to be? People almost always estimate a higher grade point average for the second question. However, both passages give the same useful information: that the student studies three hours per week. People seem to pay attention to irrelevant information in the second passage and forget that the student only studies for three hours a week. When useless information is added, it dilutes the useful information.

Gambler's Fallacy:                           Suppose you have flipped a coin and it has come up heads several straight times. The gambler's fallacy is the notion that tails is more likely to occur on the next toss. The chances of heads or tails is 50/50 on every throw. Each toss is completely independent of the other tosses. Gamblers often base their theories on the incorrect assumption that something is "bound to happen".

The Barnum Effect:                         Barnum was the guy who said, "There's a sucker born every minute". This is the tendency for people to accept favorable predictions and descriptions that are generally true of everyone, such as in horoscopes. This is a phenomenon that is quite apparent today with the popularity of the Psychic Friends Network, etc.

Overconfidence Phenomenon:       This is the tendency for people to be more confident than correct in their judgments. Studies have shown that when people answer multiple choice questions like, "Which is longer, the Mississippi River or the Nile River?", if the answers are 60% correct, the subjects are 75% certain of their answers.