Learn value, momentum, quality factors — Stock-Market notes

WHY do factors exist at all?

WHY it matters: The market return ( $R_m$ ) is one factor (market beta). But regressions show much of a portfolio's return comes from other tilts. If you can identify a repeatable tilt, you can build a strategy — and you know what you're actually being paid for.

The core model behind this is a multi-factor regression:

$R_i - R_f = \alpha_i + \beta_{m}(R_m - R_f) + \beta_{v}\,\text{VAL} + \beta_{mo}\,\text{MOM} + \beta_{q}\,\text{QUA} + \varepsilon_i$

$R_i - R_f$ = stock's excess return over the risk-free rate.
Each $\text{VAL/MOM/QUA}$ is the return of a long–short factor portfolio.
$\beta$ = how much the stock loads on that factor.
$\alpha$ = leftover skill/luck the factors cannot explain.

HOW a factor portfolio is built (the universal recipe):

Score every stock by the characteristic (e.g. cheapness).
Sort stocks into buckets (deciles/quintiles).
Go long the top bucket, go short the bottom bucket.
The long–short return = the factor premium.

Figure — Learn value, momentum, quality factors

1. VALUE — "buy cheap"

WHY does it pay? Two steel-manned stories:

Risk story: cheap firms are often distressed/fragile, so you're compensated for bearing that risk.
Behaviour story: investors over-extrapolate bad news, pushing prices too low; they mean-revert.

HOW to compute a value score — a common one uses earnings yield $E/P$ (inverse of the P/E ratio):

$\text{Value score} = \frac{E}{P} = \frac{\text{Earnings per share}}{\text{Price per share}}$

Why invert P/E? So that higher score = more attractive, matching the direction of the sort (top decile = cheapest).

2. MOMENTUM — "buy recent winners"

WHY the "12–1" window? The standard signal is the return from 12 months ago to 1 month ago:

$\text{MOM signal} = \frac{P_{t-1}}{P_{t-12}} - 1$

Why skip the last month? Very short-term (1-month) returns tend to reverse (bid–ask bounce, liquidity). Including it would contaminate the "trend" signal with noise. Skipping it keeps the clean trend.

WHY does momentum pay? Behaviour: investors under-react to news at first (prices drift up slowly), then eventually over-react (bubble → crash). Momentum harvests the drift.

3. QUALITY — "buy healthy businesses"

WHY does it pay? Behaviour: investors underpay for durable profitability, chasing exciting story-stocks instead. Risk: high-quality firms are safer, so a pure-risk model would predict lower return — yet they've historically earned more, an anomaly.

HOW — gross profitability (Novy-Marx):

$\text{GP/A} = \frac{\text{Revenue} - \text{COGS}}{\text{Total assets}}$

Why gross profit not net income? Gross profit is "cleaner" — less distorted by accounting choices, taxes, one-off charges, so it better predicts future profitability.

Combining factors (the 80/20 core)

For a 50/50 blend of two factors with returns $R_1, R_2$ :

$\text{Var}(R_p) = \tfrac14\sigma_1^2 + \tfrac14\sigma_2^2 + 2\cdot\tfrac12\cdot\tfrac12\,\rho\,\sigma_1\sigma_2$

Why this helps: if $\rho<0$ (value vs momentum), the last term is negative, so combined risk drops below the average of the two — same expected return, less volatility → higher Sharpe.

Flashcards

What is a "factor" in finance?

A systematic, persistent characteristic shared by many stocks that drives their returns beyond the market; its premium is the extra return per unit of exposure.

Universal recipe to build a factor portfolio?

Score all stocks by the characteristic, sort into buckets, go long the top bucket and short the bottom; the long–short return is the factor premium.

Why do we subtract the bottom decile from the top?

To cancel out the overall market move and isolate the pure effect of the characteristic.

Classic academic measure of value?

Book-to-market

B/M

(or equivalently earnings yield

E/P

); high = cheap = value.

Why invert P/E into E/P for scoring?

So higher score means more attractive, matching the top-decile long leg.

What is the momentum "12–1" signal?

Return from 12 months ago to 1 month ago:

P_{t-1}/P_{t-12}-1

Why skip the most recent month in momentum?

Short-term (1-month) returns tend to reverse (liquidity/bid-ask noise); skipping keeps a clean trend signal.

Behavioural story for momentum?

Under-reaction to news causes slow price drift; momentum captures the drift.

Behavioural story for value?

Over-extrapolation of bad news pushes cheap stocks too low; they mean-revert.

Novy-Marx quality measure and why gross profit?

Gross profitability GP/A = (Revenue − COGS)/Total assets; gross profit is less distorted by accounting than net income.

Why combine value + momentum + quality?

Their returns are low/negatively correlated, so blending diversifies and raises the Sharpe ratio without giving up expected return.

In the factor regression, what does alpha represent?

The return the identified factors cannot explain — leftover skill or luck.

Recall Feynman: explain to a 12-year-old

Imagine a huge shelf of toys (stocks). Instead of buying one of each, you use smart rules. Value: buy toys on sale for less than they're worth. Momentum: buy the toys everyone has started grabbing lately, because the rush keeps going a bit. Quality: buy the well-built toys that don't break. And the neat trick: use all three rules together, because on days the "on-sale" rule loses, the "everyone's grabbing it" rule often wins — so your toy chest never crashes all at once.

Connections

Fama-French Three & Five Factor Models
CAPM and Market Beta
Behavioral Finance - Over- and Under-reaction
Portfolio Variance and Diversification
Sharpe Ratio
Long-Short Equity Strategies

Concept Map

Hinglish (regional understanding)

Intuition Hinglish mein samjho

Dekho, market mein sirf "index kharido" ke alawa bhi smart tarike hote hain jinko factors kehte hain. Factor ka matlab hai ek rule jisse aap stocks ko chaanto — aur us slice ne historically extra return diya hai. Teen famous factors hain: Value (sasta kharido — jaise low P/E ya high book-to-market), Momentum (jo stocks recent 12 mahine mein bhaag rahe hain unko kharido, par last 1 month skip karo kyunki wo reverse hota hai), aur Quality (achhi, profitable, kam-karza wali companies — jaise high gross profit by assets).

Banate kaise hain? Simple recipe: sabhi stocks ko score do, deciles mein sort karo, top wale long karo aur bottom wale short karo. Top minus bottom = factor ka pure return. Bottom ko minus karne se poore market ka move cancel ho jaata hai, sirf characteristic ka asar bachta hai. Yeh important hai kyunki tab aapko pata hota hai ki aap exactly kis cheez ke liye paisa kama rahe ho.

Sabse bada gyaan (80/20): teeno ko mila do. Value aur Momentum aksar opposite direction mein chalte hain (correlation negative), isliye jab ek kharab karta hai, doosra bacha leta hai. Quality overall risk kam kar deta hai. Result — same expected return par kam volatility, yaani zyada Sharpe ratio. Isko yaad rakho: CHEAP, HOT, HEALTHY. Bas yahi factor investing ka dil hai.

Test yourself — Factor & Behavioral Finance

Connections