Study the CFA curriculum by pairing every concept with the conditions under which it applies. Write one-line decision rules for model and measure selection, drill short scenarios where you name and justify the correct tool before any calculation, and keep an error log that distinguishes concept confusion from arithmetic slips.
Choosing the Valuation Model Before You Calculate Anything
Match the model to the firm: dividend discount models for stable payers, free cash flow models for unstable payouts or shifting leverage, and relative multiples when absolute inputs are unreliable or a peer ranking is what the question asks for.
The dividend discount model assumes dividends track earnings and payout policy is stable, so it suits mature firms with predictable distributions. Free cash flow models value the firm (FCFF) or the equity (FCFE) directly from cash generation, which fits firms paying few dividends, changing their capital structure, or facing takeover. Multiples such as P/E or EV/EBITDA skip long-horizon forecasts but only rank a stock against chosen peers; they cannot establish intrinsic value, and they inherit every flaw in the benchmark set.
Worked scenario: Northline Materials reports negative net income after a one-time plant impairment, while peers trade near 18x earnings. A plausible mistake is averaging peer P/Es, declaring Northline not meaningful, or quietly substituting a pre-impairment P/E and calling the stock cheap. The better decision is to re-select the tool: first test whether the impairment is genuinely non-recurring, then compare Northline on EV/EBITDA against peers, or run an FCFF model if its operating cash flows are steady. The selection step changes the conclusion, so practicing it explicitly is the point.
Telling Apart Net Income, EBITDA, and Free Cash Flow
Each measure answers a different ownership question. Net income belongs to shareholders after every cost; EBITDA belongs to all capital providers before depreciation and amortization; FCFF values the whole firm; FCFE values equity after debt flows.
The differences matter because each measure must pair with a matching value in the denominator or the discounting. Equity measures (net income, FCFE) pair with market capitalization or equity value; firm-wide measures (EBITDA, FCFF) pair with enterprise value. Mixing them silently distorts every conclusion. Capital-structure differences are the practical driver: two identical businesses with different debt loads show very different net income but nearly identical EBITDA and FCFF, which is why cross-company comparisons so often require the firm-wide measures.
Apply this to a leveraged retailer: interest expense crushes net income, so its P/E looks expensive next to unlevered peers, and a mistake would be concluding the shares are overpriced on that basis. EBITDA neutralizes the financing difference, and FCFF goes further by counting interest as a flow to debt holders rather than a cost. FCFE behaves differently again: it shrinks when the firm repays debt, even though the business is healthy. Naming which flows belong to which claimants is the concept to rehearse until it is automatic.
The table below compresses the pairing rules into one reference you can rebuild from memory.
| Measure | Question it answers | Best fit | Consistency check |
|---|---|---|---|
| Net income (P/E) | What do equity holders earn after all costs? | Profitable firms with broadly comparable payout and leverage | Pair only with equity (market) value |
| EBITDA (EV/EBITDA) | What do all capital providers earn before D&A? | Capital-intensive or differently leveraged peer groups | Pair only with enterprise value |
| FCFF | What cash is available to all investors? | Changing leverage, no or irregular dividends | Discount at WACC; subtract debt for equity value |
| FCFE | What cash is available to shareholders? | Stable debt policy, non-dividend payers | Discount at the cost of equity; no debt adjustment needed |
Sharpe, Treynor, or Jensen's Alpha: What the Scenario Implies
The correct measure depends on the portfolio's role for that client. Sharpe uses total risk and fits a client's entire wealth; Treynor and Jensen's alpha use beta and fit one sleeve of an already diversified portfolio.
The Sharpe ratio divides excess return by total volatility, so it penalizes any risk the client actually bears. Treynor's ratio divides excess return by beta, crediting diversification the investor is assumed to hold elsewhere, and Jensen's alpha compares realized return with the return the capital asset pricing model predicts for that beta. These are not interchangeable rankings of the same thing; they answer different questions about whose risk is being compensated.
Worked scenario: Portfolio A returned 10% with standard deviation 15% and beta 0.8; Portfolio B returned 9% with standard deviation 10% and beta 1.1; the risk-free rate is 2%. Sharpe ranks B first (0.70 versus 0.53) because B earns more per unit of total risk. Treynor reverses the ranking: A scores 10.0 against B's 6.4, because A delivers strong reward per unit of market risk. The mistake to avoid is defaulting to Sharpe as the standard answer. Ask instead: is this the client's whole portfolio, or one holding inside diversified wealth? That single question flips the ranking, which is why the identification step earns its place in your notes.
Applying the Code of Ethics to Situations, Not Reciting Standards
Ethics questions reward a repeatable process: establish the facts, identify the duty owed and to whom, match it to the relevant standard, and select the action the standard requires, rather than merely recognizing a rule's name.
The Code and Standards govern conduct toward clients, prospects, the market, the employer, and the profession, and a single scenario can implicate more than one duty. A reliable method is to write, in order: the parties and their relationships, the specific information or action at issue, the duty that information triggers, and the minimum action required. Practicing this four-step narration turns ethics from pattern-matching labels into a procedure you can execute on any unfamiliar fact pattern.
Scenario: a friend at a competing firm asks how your largest client is positioned, noting she recognized the client's name in a news headline. A tempting mistake is to share vague color while nodding toward the client's known concentration, assuming public coverage dissolves the duty of confidentiality. The better decision treats client information as protected regardless of how a third party learned a fragment: decline specifics, explain in principle that professional standards bar disclosure without client consent or a legal requirement, and offer nothing. Notice that the graded action is what you do and say, not the standard's title, so rehearse the action sentence itself.
Reading Statements for Earnings Quality, Not Just Ratios
Financial assessment asks whether reported earnings represent durable cash generation. Compare accruals to operating cash flow, strip out non-recurring items, and check revenue recognition timing before trusting any ratio built on net income.
Accrual accounting lets reported income diverge from cash collected, so a core interpretive habit is side-by-side reading of net income and cash flow from operations. Persistent gaps, receivables growing faster than revenue, and income propped up by one-time gains or estimated accruals each suggest that headline earnings may overstate repeatable performance. This connects directly to valuation: a P/E computed on low-quality earnings is precise arithmetic applied to the wrong number.
Scenario: a software company reports rising net income while operating cash flow stays flat and receivables climb sharply. The mistake is to accept the improving P/E story at face value. The better decision is to restate the earnings base: remove non-recurring items, test whether revenue recognized outpaces cash collected, and where cash backing is doubtful, re-anchor the analysis on FCFE or on cash-supported recurring earnings. The conclusion may survive or not, but the assessment you perform, and being able to state which adjustments you made and why, is the skill the curriculum is teaching.
A Self-Check Exercise: Justify the Method Before the Math
Take short company or portfolio profiles and, in writing, choose a valuation model or performance measure and defend it in two sentences. Grade the justification, not the arithmetic, using the rubric below.
Build ten one-paragraph profiles that deliberately mix cues: a mature dividend payer, a leveraged turnarounds candidate, a fund that is one holding in a diversified account, a firm with receivables outrunning revenue, a confidentiality-adjacent conversation with a third party. For each, give yourself roughly ninety seconds to write three things: the two or three facts that matter, the tool you would use, and one limitation or alternative. No calculation is allowed at this stage, which keeps the focus on selection, where a wrong first step poisons everything downstream.
After ten profiles, read your answers as data. You should observe clear cue-to-tool patterns emerging, for example payout stability triggering the dividend model, and you should notice exactly which cues made you hesitate. Hesitation is the diagnostic: it marks a concept to reread at the source, not more drill volume to add. Re-run the exercise a week later with reshuffled profiles and confirm the hesitations have moved to a shorter list.
- 2 points: firm or portfolio traits named explicitly (payout stability, direction of leverage change, the portfolio's role for its owner)
- 2 points: the chosen model or measure is justified by those traits, not by habit
- 1 point: one limitation or alternative tool acknowledged
- Score 5/5 consistently before moving to timed mixed practice; treat the score as a learning milestone, not a prediction of exam performance
A Six-Week Sequence From Decision Rules to Mixed Scenarios
Sequence preparation in three phases: build concept-decision maps, drill single-topic scenarios, then mix topics with labels removed. Close with an error log review that targets concept confusion first and arithmetic slips second.
The order matters because mixing topics too early produces exactly the failure this guide targets: swapping methods under pressure. Phase one builds a one-page decision map per domain, so every formula you learn is attached to its applicability conditions. Phase two drills single-topic scenarios against the Section 6 rubric, keeping difficulty within one concept so selection becomes fluent. Only phase three removes topic labels, forcing the identification step on every item, which is the condition exam-style questions actually create.
Readiness is concrete, not a feeling. You are approaching exam-style readiness when you can rebuild each decision map from memory, complete a mixed drill of twenty unlabeled items with at most two method-selection errors, and narrate an ethics scenario's required action before naming the standard involved. Log every selection error with its cause and reconcile against your maps; shrinking error categories over two consecutive sessions is the milestone that says the sequence worked. Administrative details such as registration, fees, and dates live with the issuer at cfainstitute.org and should be confirmed there.
- Weeks 1-2: write one-page decision maps (DDM vs FCF vs multiples; Sharpe vs Treynor vs Jensen's alpha; confidentiality and duty hierarchies)
- Weeks 3-4: single-topic scenario drills graded with the Section 6 rubric; log every method-selection error with its cause
- Week 5: mixed practice with topic labels removed; your first written line on every item is which tool applies and why
- Week 6: timed mixed sets, then rebuild all decision maps from memory and reconcile them against your original notes
References and further reading
Use these references to explore the concepts and check the latest information from the relevant organizations.