Learn about behavioral finance biases — Hinglish notes

6.6.6 · Stock-Market › Factor & Behavioral Finance

Behavioral biases KIYA hain?

Do broad families hain:

Cognitive biases — reasoning/information-processing mein galtiyan (faulty logic, bad statistics). Aksar education/data se theek ho sakti hain.
Emotional biases — feelings se driven galtiyan (dar, pachtawa, pride). Theek karna mushkil hota hai; inhe argue karke nahi, manage karke handle karna padta hai.

Biases KYUN exist karti hain? (First-principles derivation)

Tera brain evolve hua hai taki uncertainty mein limited energy ke saath fast decisions le sake. Isliye wo heuristics (rules of thumb) use karta hai. Ek heuristic savanna mein efficient hoti hai lekin financial markets mein misfire karti hai — yahan environment randomness, feedback loops, aur low signal-to-noise se bhara hota hai.

Key biases (WHY ye sahi lagte hain ke saath)

1. Overconfidence

Kaise dikhta hai: excessive trading, under-diversification, tight (bahut narrow) forecast ranges.
Cost: zyada trading → zyada fees + bura timing → lower net returns.

2. Anchoring

Example: "Stock pichle saal ₹1000 tha, toh ₹600 sasta hai" — tumne ₹1000 par anchor kar liya, yeh bhool ke ki fundamentals badal gaye.

3. Loss Aversion (Prospect Theory)

Isse disposition effect hota hai: winners bahut jaldi bech do, losers bahut der tak rokho (loss realize karne se bachne ke liye).

4. Confirmation Bias

5. Herding

6. Availability Bias

7. Recency Bias

8. Mental Accounting

Ek number jo derive karne layak hai: Loss Aversion coefficient

Prospect Theory smooth utility curve ki jagah ek value function use karta hai jo reference point $r$ par kinked hoti hai (usually tumhara purchase price ya current wealth):

v (x) = {(x - r)^{α} - λ (r - x)^{β} x \geq r (gains, concave) x < r (losses, convex, steeper)

"Fair coin flip nahi lunga" result derive karo. Offer: $G$ jito ya $L$ haro 50/50 mein. Ek rational risk-neutral person accept karta hai agar $G \geq L$ . Lekin loss aversion ke under (simplicity ke liye $α = β = 1$ lo, reference $0$ par):

Accept if \frac{1}{2} G - \frac{1}{2} λ L \geq 0 ⟹ G \geq λ L

Toh $λ = 2$ ke saath, tum demand karte ho ki jo jeet sako wo kam se kam do baar ho jo haar sako — sirf ek fair coin flip ke liye. Woh ek inequality disposition effect, insurance over-buying, aur crashes ke waqt "freezing" ko explain karti hai.

Figure — Learn about behavioral finance biases

Worked examples

Common mistakes (Steel-manned)

Recall Feynman: ek 12-saal ke bacche ko samjhao

Socho tumhare brain mein ek "quick-answer" button hai taki tumhe har cheez ke baare mein mushkil se sochna na pade — ball se bachne ke liye yeh great hai, lekin paison ke liye bura hai. Ek quirk: haarna jeetne se do baar bura lagta hai. Toh agar main tumhe coin flip offer karun jahan tum ₹10 jito ya ₹10 haro, tum kahoge "nahi thanks," even though yeh bilkul fair hai. Stock market mein isse log buri stocks rok lete hain (break even ki umeed mein) aur acchi stocks bahut jaldi bech dete hain. Trick jaanna tumhe khud ko yeh karte hue pakadne deta hai.

Flashcards

Ek error ko "bias" kya banata hai, random noise nahi?

Yeh systematic hota hai — yeh poori crowd ko predictably ek hi direction mein dhakelta hai, toh cancel nahi hota aur prices move kar sakta hai.

Cognitive vs emotional bias?

Cognitive = faulty reasoning/statistics (data se theek ho sakta hai); Emotional = feelings jaise dar/pachtawa/pride se driven (manage karna padta hai).

Disposition effect?

Loss realize karne se bachne ke liye winners bahut jaldi bechne aur losers bahut der tak rokne ki tendency.

Typical empirical loss-aversion coefficient λ?

Lagbhag 2.25 — losses ~2.25× barabar gains se zyada intense lagte hain.

Loss aversion ke under (α=β=1), fair 50/50 win-G/lose-L bet kab accept karte ho?

Tabhi jab

G \geq λ L

, matlab potential win, potential loss ke λ times se zyada hona chahiye.

Ek loss-averse person risk-SEEKING kyun ho sakta hai?

Value function loss domain mein convex hai, toh wo "get back to even" karne ke liye gamble karte hain.

Anchoring bias?

Value estimate karte waqt pehle dekhe gaye number par zyada rely karna.

Availability vs recency bias?

Availability = recall ki aasaani se probability judge karo; recency = forecasts mein most recent events ko zyada weight do.

Mental accounting error?

Paisa arbitrary label ke basis par differently treat karna, yeh ignore karte hue ki paisa fungible hota hai.

Overconfidence returns kyun kam karta hai?

Isse over-trading aur under-diversification hoti hai → zyada costs aur kharaab risk-adjusted returns.

Connections

Prospect Theory — loss aversion ka formal model
Efficient Market Hypothesis — rational benchmark jise biases violate karti hain
Factor Investing — kuch factors (momentum, value) bias-driven mispricing ke liye paid premia ho sakte hain
Limits to Arbitrage — biases instantly correct kyun nahi hoti
Risk Aversion vs Loss Aversion
Herding and Market Bubbles