Lecture 14 Biases

Biases
- A cognitive, emotional distortion in DM
- We cannot explain why we perceive the $\dfrac{\text{Length}}{Breadth}$ ratio to be different for: these two tables. ¹
Two kinds of System
- "Why are we good at certain things, but dumb at others? Why this contradiction?"
- System 1: Intuitive & Automatic system
  - rapid and feels instinctive, doesn't involve conscious thinking
  - associated with the oldest parts of the brain (we share with lizards)
- System 2: Reflective & rational system
  - deliberate and self-concious

Automatic System	Reflective System
Uncontrolled	Controlled
Effortless	Effortful
Associative	Deductive
Fast	Slow
Skilled	Rule-following
- We can train our automatic system to be a reflective one.
- Playing games (National level: Automatic, "trying in Mundus": Reflective)
- Instruments (Maestro: Automatic, Grade 1: Reflective)
- Chess (Master: Automatic, Novice: Reflective)
- Language (Native: Automatic, Foreign: Reflective)
- Examples of Intuitive Thinking
- Bat and ball cost $1.10, Bat cost $1.00 more than ball. Ball costs $\_\_\_\_$? (not 10 cents)
- 5 machines make 5 widgets in 5 minutes. 100 machines make 100 widgets in $\_\_\_\_\_$? (not 100)
- Path of lily pads. Every day, doubles in size. Takes 48 days to cover entire lake. To cover half the lake it takes $\_\_\_\_\_$? (not 24)
- Shane Frederick #economist : Most popular answers even among bright college students.
- Econs (unlike humans) never make an important decision without checking their reflective system.

1 More on Biases¶

Optimism & Overconfidence
- Unrealistic optimism is a pervasive feature² $\implies$ Risk taking behaviors (especially in life and health)
- Overconfidence: perception/judgement about ability > actual objective outcome. Defined in 3 ways:
  1. Overestimation of actual performance
  2. Over-placement of performance, relative to others
  3. Over-precision in expressing unwarranted certainty in accuracy of beliefs
- e.g.
  - student's performance
  - confidence in one's profession incl. entrepreneurship
  - success in marriage³
- Unrealistic optimism explains risk-taking in life and health:
  - overestimate their personal immunity $\implies$ failure to make sensible preventive steps
  - They might benefit from a nudge
Loss aversion ^669922
- Psychologically, $Pain(\text{Loss}) > Pleasure(\text{Gain})$ even if $\text{Loss} = \text{Gain}$ objectively.
- $\implies$ Inertia to hold on (stick) with your current belongings (holdings). "I don't want to depart from the thing that I have."
- Experiment with Mugs trading
  - Half the students (owners) given: Mugs with insignia of their home university
  - Other half (non-owners) asked to examine the mugs
  - Owners are invited to sell. Non-owners to buy
  - Owners demanded (roughly) twice as much to give up their mugs as non-owners were willing to pay
- $\implies$People don't assign specific (objective) to objects. When they have to give up something, they are hurt more than they are pleased if they acquire the same thing.
- Another experiment with mugs
  - Half: Coffee mugs, other half: chocolate bars (Same value)
  - When asked, only 1 out of 10 switched
- Example with bets
  - Heads: Win $$X$.
  - Tails: Lose $100 (out of their pocket)
  - "How much does $X$ need to be for you to take the bet?"
  - Most people answered: $X =$ $200.
  - $\implies$ Prospect of winning $200 just offsets the prospect of losing $100.
Disposition effect
- Coined by Hersh Shefrin and Meir Statman (1985) #economist
- DE describes behavior of investors when selling (disposing off) their winning position too soon or holding a losing position too long.
- Rational decision making suggests:
  - Sell "losers" for tax reductions
  - Decision for shorting or longing should depend on $E(\text{Future Value})$ and not on purchase $\text{Price}$.
- DE is driven by expected feelings of
  - Pride: making the right decision: selling at a gain
  - Regret: making the wrong decision: selling at a loss
Illusion of control
- Belief that we have control or influence over some events, while the actual events are beyond our control or understanding.⁴
- Traditional theories: Humans have strong motives to have control over life and environment
- Studies: Controllability Dimension is considered an achievement.
  - Failure of this measure $\Leftarrow$ personal causes
- IOC is a defense mechanism $\sim$ self-esteem for an individual/nation.
- Examples:
  - 2008 Financial Crisis: Risk management in the banking sector $\implies$ 2008 financial crisis. IOC encourage more risk-taking until it became uncontrollable.⁵
  - Gamblers throw the dice harder to get higher numbers (belief: with "skill" they can somehow control fortune)
  - Participants in a lottery experiment: "If I choose the numbers, I have more control over the outcome."
- IOC can also foster mental health: people receive satisfaction by exerting some level of control in their life.
  - Motivates them to be persistent rather than giving up
  - Studies: Adaptive effect of optimistic beliefs #doubt
The hindsight bias (HB)
- "Knew-it-all-along" or creeping determination.
- Tendency to overestimate the ability to have predicted an outcome. OR Belief that one actually predicted it correctly, after the outcome is known.
- $\Leftarrow$ False Memory, anchoring, overconfidence, personal motivation
- Pierre Janet and Sigmund Freud #psychologist: Study of false memory.
  - In severe cases, the person recalls "a memory" that didn't actually happen
  - Less severe cases, distorted foresight predictions (which they didn't actually do)
  - e.g. I think I will fail the exam, but actually pass and then I say... "I knew I could pass it".
- HB is revealed through:
  - multifactor deficiencies & traits of limitations in humans
    - e.g. Selective memory retrieval, anchoring, illusion of foreseeability
    - these are representations of self-fulfilling prophecy #doubt
  - e.g. Anchoring in HB: Investor reads predictions about gold price movements. Then, invests. If prices actually go up, he says "I knew I have predicted it right that it would go up!"⁶
The conjunction effect/fallacy (CF)
- Kahneman and Tversky in their experiment: "Conjunction is a fallacy"
  - Let $A$ and $B$ be two things
  - $P(A) \ge P(A\cap B) \le P(A\cup B)$
- Their 1983 experiment
  - Bill is 34 years old, intelligent... unimaginative, compulsive and generally lifeless
  - Strong in Mathematics, weak in Social Studies & Humanities
  - Rank from most to least probable: ⁷
  - Observations
    - 92% of 94 undergrads gave $A \gt E \gt C$
      - i.e. P(Accountant) > P(Accountant & plays Jazz) > P(Plays jazz for a hobby)
      - Really, $E = A \cup C$ $\implies$ the students really said $P(A) \gt P(A\cup C) \gt P(C)$ which violates the basic rule mentioned before.
    - 83% of 32 graduate students: $E \gt C$ in the decision science program of Stanford Business School (with advanced courses in probability and statistics)
  - Researchers say, "It's wrong to impose rules of probability as a norm for making rational inferences under uncertainty"
    - $\implies$ Intelligent inferences were mistaken as reasoning errors
- Series of studies:
  - If the word "probability" is used, people consider the non-mathematical meaning of it
  - If, instead, "frequency" is used, possible interpretations of "probability" is narrowed down to its mathematical meaning.
- Testing the CF problem using frequencies
  - Tversky and Kahneman rephrased as. How many of them are:
    - Accountants (out of 100)?
    - Accountants and play jazz for a hobby (out of 100)?
  - $\implies$No conjunction fallacy observed!

When in reality, the lengths and breadths of the two are exactly the same. ↩
Pervasive means wide-spread ↩
Since no prior information available, the odds of success is 50%. Yet, we have the optimism that it would work out. Most people dating experience breakup, because we are optimistic that our heart is not going to be broken. ↩
Now when I reflect about it, when I see some of my band juniors fretting about something that they need not. I too have fretted about things that Shyam sir told me that I need not. Some things are just beyond our control, no matter how much our mind tells us they are. ↩
Banks created subprime mortgages at lower interest rates to many people for buying houses. They didn't check their credit history, didn't verify their income. this led to many high-risk loans. Financial institutions, in order to spread the risk created Mortgage-backed-Securities (MBS) and Collateralized-Debt Obligations (CDOs) and many investors bought these. In 2006, the Housing Bubble Burst, house prices started to fall. Interest rates adjusted upwards $\implies$ Unable to pay back the mortgages and defaulted. House prices fell, interests raised $\implies$ MBS and CDOs collapsed. Nobody lent to each other, widespread credit crisis. Lehman Brothers collapsed and the stock market crashed. There was a great recession $\implies$ Widespread job loss, unemployment and decline in economic activity. ↩
Bugger, you read it somewhere. ↩
I ranked it as: C > B > F > H > E > D > G > A ↩