AMAZON COM INC

What it measures · Benjamin Graham's conservative safety-margin formula. A high value requires both earnings and equity to be solid, so read it as 'the ceiling for a financially intact company.'

How it is computed · √(22.5 × NetIncome × Equity). 22.5 = max P/E 15 × max P/B 1.5 Graham would pay. Net income is latest FY consolidated net income from DART/XBRL; equity is total equity from the same filing.

When to trust it · Best for stable profitable industrials. Underestimates SaaS/platform/biotech because intangible-heavy businesses carry little book equity.

What it measures · Graham's 1962 revision that bakes growth into the multiple. Unlike the static Graham Number, this scales linearly with EPS growth so growth stocks aren't punished as hard.

How it is computed · NetIncome × (8.5 + 2g). 8.5 = base P/E for a zero-growth firm; 2g = +2× P/E for every 1pp of EPS growth. g is YoY EPS growth, clamped to [-5%, +25%] so a single big year doesn't blow up the estimate.

When to trust it · Works for firms with proven, somewhat stable growth. Misleading for first-year profits, swing years, or any period where g is distorted by a one-time charge or gain.

What it measures · Relative-value benchmark — "should trade at the industry average P/E." Not an absolute intrinsic value; tells you how the firm prices vs its peer set.

How it is computed · NetIncome × industry P/E. Defaults to a market-wide 15× when no industry multiple is provided. Net income is the latest FY consolidated net income.

When to trust it · Most reliable in mature industries with comparable peers (banks, telcos, utilities, refiners). Less useful for big-tech / hardware where industry P/Es span 10×–60×.

What it measures · Multiplies book equity by industry P/B to imply a fair market cap. Best for asset-heavy businesses (banks, insurers, brokers, REITs) where balance-sheet value is the franchise.

How it is computed · Total Equity × industry P/B. Defaults to 1.5×. Equity is total equity (or parent-owners share) from the latest balance sheet.

When to trust it · Strongest for banks/insurers — asset-liability matching is the business. Materially under-prices SaaS, platforms, and brand-heavy businesses where intangible value dwarfs book equity.

What it measures · Multiplies revenue by industry P/S. Useful when earnings are negative or noisy (early-stage, restructuring) — revenue stays positive.

How it is computed · Revenue × industry P/S. Defaults to 2.0×. Revenue is the latest FY consolidated top line.

When to trust it · Ignores margin structure — a 30% operating margin business and a -5% one get the same multiple. Only meaningful within peer + margin cohort.

What it measures · Simplified Gordon-growth DCF — discounting perpetual cash flows. Extremely sensitive to the cash-flow / discount-rate / growth-rate inputs; trust the scenario grid more than the headline number.

How it is computed · FCF × (1+g) / (r − g). FCF preferred from XBRL/DART; falls back to NetIncome, then 5% of Revenue (rough proxy). Discount rate r = risk-free (4%) + equity premium (5%) = 9%. Growth g is EPS YoY clamped to [-2%, +6%], default 3%.

When to trust it · Strongest for steady, predictable businesses. When r-g approaches 0, the denominator blows up; near-perpetuity-growth combos are intentionally voided. Useless for cyclicals (semis, commodities, shipping) where g itself is noise.

What it measures · Values the firm purely on dividends paid to shareholders. Cleanest in theory, but only meaningful when management clearly intends to pay dividends — useless for share-repurchase-only payout policies.

How it is computed · D × (1+g) / (r − g). D from DART alotMatter (배당사항) or |CF dividendsPaid|. If absent, cascades to 30% of NetIncome (typical payout) → 30% of FCF → 1.5% of Revenue. r and g assumptions match DCF.

When to trust it · Most reliable for stable high-yielders (Korean utilities, tobacco, financial holdings). Sharply under-prices buyback-heavy firms (US big tech, parts of Korean IT) — look at the TSY column on the listing page to balance.