What This Volume Adds: The Escalation Map, How to Read It and the Quant Toolkit

Volume I answered a structural question: how money, banking, markets, industry, and policy fit together into one system. By its end, you could trace a Federal Reserve rate decision all the way through to a reconciliation break on a trading desk. That is financial literacy at the level of the whole system, and it is valuable — but it is the understanding of an informed observer, not yet the working vocabulary of a practitioner.

This volume closes that gap. It takes the nouns Volume I gave you — EBITDA, market capitalization, the yield curve, a board of directors — and turns them into verbs: how you actually build a valuation, structure a deal, price a derivative, measure portfolio risk, and run a board-level CEO search. Where Volume I would tell you that P/E is a valuation ratio, this volume shows you how to build the full discounted cash flow model that a P/E multiple is really a shorthand for. Every Part contains at least one fully worked numeric example, and the statistics those examples rely on (expected value, volatility, correlation, regression, simulation and rate conventions) are collected once, in the Quant Toolkit (Section 0.4: The Quant Toolkit).

Nothing in this volume starts from zero. Every Part below is a direct escalation of something Volume I already introduced at an observer’s level of depth.

Part 1: Corporate Finance & Valuation is the right starting point for everyone, since valuation mechanics underpin almost everything that follows — you cannot structure an LBO (Part 2: M&A, Private Equity & LBOs) or run a comparable company analysis (also Part 1: Corporate Finance & Valuation) without first being fluent in discounted cash flow and cost of capital. After Part 1: Corporate Finance & Valuation, the remaining Parts are largely independent of one another and can be read in whatever order matches your immediate need. If governance and executive search are your priority, Part 7: Governance & Executive Search can be read directly after Part 1 without loss of continuity.

Read the Quant Toolkit (Section 0.4: The Quant Toolkit) before Parts 3 to 5, or keep it open beside them: bond duration, option pricing, portfolio variance, value at risk and factor models all use its seven tools, and each tool there points forward to the section that uses it. Part 12: The Full Stitch II draws on the Parts before it: Case One reruns the Part 2 LBO, Case Two revalues a company on restated earnings (Parts 1 and 6), and Case Three prices a currency hedge (Part 4: Derivatives, Properly Priced).

Volume I taught the system without statistics; Parts 3 to 5 cannot be read without them. Seven tools carry the load, each shown once with a worked number (illustrative inputs, exact arithmetic, rounded for display) and a pointer to where it is used.

Expected value is the probability-weighted average of outcomes: E[X] = Σ pᵢ × xᵢ. A product launch has three outcomes: recession (25%, −$600,000), base (50%, +$200,000), boom (25%, +$900,000). E[X] = −150,000 + 100,000 + 225,000 = $175,000, a payoff no single scenario delivers: it is the long-run average across many such bets. It reappears as expected credit loss, probability of default × loss given default (Part 3.6: Credit Spreads Decomposed), and in option pricing (Part 4.4: Black-Scholes — The Intuition Behind the Formula).

Variance measures scatter around the mean: σ² = Σ pᵢ × (xᵢ − E[X])². For the base launch, 0.25 × (−775,000)² + 0.50 × 25,000² + 0.25 × 725,000² = 281,875,000,000 squared dollars. Its square root, the standard deviation, is back in dollars: $530,919. For returns it is called volatility, and with independent days it scales by the square-root-of-time rule: 1.0% daily × √252 = 15.9% a year. It drives the Sharpe ratio (Part 5.4: The Sharpe Ratio & Risk-Adjusted Return), value at risk (Part 5.5: Value at Risk (VaR) — Measuring Downside Risk) and Black-Scholes (Part 4.4: Black-Scholes — The Intuition Behind the Formula).

Covariance measures whether two returns move together; dividing it by both standard deviations gives correlation, ρ, between −1 and +1. Portfolio variance is σ_p² = w_A²σ_A² + w_B²σ_B² + 2w_Aw_Bρσ_Aσ_B. For 60% stocks (σ 18%) and 40% bonds (σ 6%) at ρ = −0.2: 0.011664 + 0.000576 − 0.0010368 = 0.0112032, so σ_p = 10.58%, against 13.2% if ρ were +1. That 2.6-point gap is diversification (Part 5.1: Diversification — Why Combining Assets Reduces Risk, Part 5.2: Modern Portfolio Theory & the Efficient Frontier).

The normal distribution is the bell curve set by a mean and a standard deviation; a z-score, z = (x − μ) ÷ σ, counts standard deviations from the mean. Under normality 68.3% of outcomes lie within ±1σ and 95.4% within ±2σ, and the one-tailed 99% cutoff is z = 2.326. Parametric VaR is exactly this: a $10,000,000 portfolio with daily σ of 1.2% has 1-day 99% VaR = 2.326 × 0.012 × $10,000,000 = $279,120 (Part 5.5: Value at Risk (VaR) — Measuring Downside Risk). The weakness is the tail. On October 19, 1987 the Dow fell 22.6%; at an illustrative 1% daily σ that is z = −22.6, with a normal probability near 10⁻¹¹³. Real returns have fat tails (Part 5.6: VaR’s Limitations — What It Doesn’t Capture).

Regression fits the line that minimizes squared errors. Regress a stock’s returns on the market’s and the slope is beta: β = ρ × σ_s ÷ σ_m = 0.6 × 30% ÷ 15% = 1.2, the beta of Part 1.5: CAPM — The Cost of Equity. R-squared = ρ² = 0.36: the market explains 36% of the stock’s variance; the other 64% is diversifiable. With 60 monthly observations the standard error is √[(1 − R²) ÷ (n − 2)] × σ_s ÷ σ_m = √(0.64 ÷ 58) × 2 = 0.21, so the 95% range is 0.79 to 1.61. Because betas drift toward 1 (Blume, 1975), vendors often publish an adjusted beta, 0.67 × 1.2 + 0.33 = 1.13. More regressors give the factor models of Part 5.8: Factor Investing — Fama-French and Beyond.

Monte Carlo simulation draws thousands of random scenarios from assumed distributions and reads the answer off the results (Boyle, 1977, brought it to option pricing). Simulate $1,000,000 over 10 years with annual returns drawn from a normal with mean 7% and σ 15%, 100,000 paths. The mean ending value, about $1.97 million, matches the formula $1,000,000 × 1.07¹⁰ = $1,967,151, the check that the model is wired correctly. The median is only about $1.79 million and about 10% of paths lose money. The gap is volatility drag: the typical path compounds near μ − σ²/2 = 5.9%. Part 5.7: Monte Carlo Simulation — Modeling Many Futures applies the method to VaR.

Compounding and day-count conventions decide what a rate means. The effective annual rate (EAR) = (1 + APR ÷ m)^m − 1: a 6% APR compounded monthly is 6.168%, and with continuous compounding, e^0.06 − 1 = 6.184%. US deposits quote the annual percentage yield (APY), defined by Regulation DD on a 365-day basis. The day-count convention sets the year: SOFR uses actual/360; municipal bonds use 30/360. On $10,000,000 at 5% for 90 days, actual/360 pays $125,000 and actual/365 pays $123,288: 5% actual/360 is 5.069% on a 365-day basis. These sit under Part 1.2: The Time Value of Money — The Idea Underneath Everything, Part 3.1: Bond Pricing Basics — Price, Yield, and the Inverse Relationship, Part 3.4: The Yield Curve, Mathematically — Spot Rates and Forward Rates, Part 4.1: Futures & Forwards — Locking In a Price and Part 4.5: Interest Rate Swaps.