Risk and compliance interview questions: the 92 topics they test

Technical questions in a risk and compliance interview come back to these 92 topics, from 13 guides of the Academy. Each one is the rule to know in one sentence, linked to its guide: the why, a worked example and a timed quiz are there.

Market and credit risk

Measure risk like a bank risk team: VaR and its blind spots, backtesting, expected shortfall and stress tests, counterparty exposure and CVA, expected loss and the other risks that sink banks.

  1. Computing VaR: parametric, historical and combined. A normal VaR is the multiplier × volatility × value, 1.65 at 95% and 2.33 at 99%, and it is a threshold, not the largest possible loss.
  2. Where VaR fails: tails, liquidity, options and backtesting. A normal-distribution VaR understates fat tails, assumes positions can be sold within the horizon and misses the non-linear losses of short options.
  3. Expected shortfall, stressed VaR and stress tests. Expected shortfall averages losses beyond the VaR; Basel uses it at 97.5%, close to a 99% VaR under normality and larger with fat tails.
  4. Counterparty risk: exposure, netting, CVA and wrong-way risk. Counterparty exposure is today's positive value plus potential future exposure minus collateral, and netting offsets trades with the same counterparty.
  5. Expected loss, capital and concentration. Expected loss is probability of default × loss given default × exposure at default, covered by pricing and provisions; capital absorbs unexpected losses.
  6. Operational, model, basis and banking-book risks. Rogue trading is operational risk, from failed people, processes and controls, while model risk is the loss from a wrong or misused model.
  7. When VaR punishes diversification. VaR can rise when 2 risks are combined, so it can penalise diversification, while expected shortfall never exceeds the sum of the parts.

Compliance, KYC and AML

Think like a compliance officer: laundering stages and red flags, KYC and beneficial owners, PEPs and sanctions, market abuse and information barriers, and protected whistleblowers.

  1. Money laundering: stages, red flags and reporting. Laundering runs through placement of cash, layering that hides its origin and integration as apparently clean wealth.
  2. KYC: beneficial owners, risk-based checks and reviews. Under EU rules anyone holding more than 25% of shares or votes, or control by other means, is a beneficial owner to identify.
  3. Red flags: PEPs, sanctions and shell companies. A politically exposed person and their close family get enhanced due diligence, kept for at least 12 months after leaving office.
  4. Market abuse: insider dealing, spoofing and wash trading. Trading on inside information, such as an unannounced takeover, or passing it on, is insider dealing, banned under MAR.
  5. Information barriers, lists, personal trading and records. Information barriers keep private-side inside information, such as M&A, from public-side sales, trading and research staff.
  6. Whistleblowing, conflicts and bribery. A whistleblowing report must be protected and investigated, and retaliation is itself a serious breach.
  7. Ownership chains: multiply along, add across. An indirect stake is the product of the percentages along each chain of ownership, and stakes through different chains add up.

Statistics and machine learning

The statistics quant interviews lean on: regression and its traps, correlation, what significance does and does not mean, normal tails, return arithmetic and the basics of model fitting.

  1. Regression and its traps. The OLS slope of y on x is the correlation times σy divided by σx, and with one regressor R² is the squared correlation.
  2. Correlation, variance of sums and spurious links. The variance of a sum adds the variances and twice the covariance, and correlations of trending price levels are often spurious, so correlate returns.
  3. Significance, errors and intervals. A p-value is the probability of data at least this extreme if the null hypothesis is true, and a mean's standard error is σ ÷ √n.
  4. Normal quantiles, skewness and fat tails. For a normal variable about 68.3% lies within 1 standard deviation and 95.4% within 2, and 1.645 is the one-sided 5% quantile.
  5. Return arithmetic and robust statistics. A Sharpe ratio scales with the square root of time, because mean returns grow with time and volatility with its root.
  6. Fitting models: overfitting, regularisation, metrics. A model that fits training data far better than new data has overfitted, and Lasso can set coefficients to zero while ridge only shrinks them.
  7. The Sharpe ratio luck alone would find. A Sharpe ratio estimated over T years has a standard error of about the square root of 1 ÷ T for a true Sharpe near zero.

Credit risk

Work like a credit risk analyst: IFRS 9 stages and expected credit loss, PD and LGD models, collateral and exposure at default, portfolio concentration and watchlists, and loan pricing on risk.

  1. IFRS 9: stages, expected credit loss and scenarios. Under IFRS 9 a stage 1 loan carries 12-month expected credit loss, and stage 2, after a significant increase in credit risk, lifetime loss.
  2. PD and LGD models: default, cycles, ranking and calibration. Default means more than 90 days past due or unlikely to pay, and through-the-cycle (TTC) PDs span a cycle while point-in-time PDs reflect today.
  3. Collateral, haircuts and exposure at default. Collateral counts at its value after a haircut, larger for volatile assets such as listed equities, so LGD is the uncovered share of the exposure.
  4. Portfolio quality: NPLs, concentration, early warning and forbearance. The NPL ratio is non-performing loans ÷ total loans, and coverage is provisions ÷ non-performing loans.
  5. Pricing loans on risk and transferring it. A loan's minimum rate adds funding, expected loss, the cost of the capital it uses and running costs.
  6. Same expected loss, very different tails. Default correlation leaves expected loss unchanged but widens the spread of losses, so capital, sized on unexpected loss, depends on correlation.

Banking regulation

Read a bank's capital and liquidity like a supervisor: CET1 and buffers, risk weights and the output floor, leverage, LCR and NSFR, the 3 pillars, AT1 and bail-in.

  1. CET1, the buffer stack and deductions. The CET1 ratio is CET1 ÷ risk-weighted assets, and Basel III requires a 4.5% minimum plus a 2.5% conservation buffer.
  2. Risk weights, internal models and the output floor. Under Basel, capital needed is risk weight × exposure × the required ratio, and mortgages weigh less than unsecured corporate loans because property secures them.
  3. Leverage ratio, LCR and NSFR. The leverage ratio is Tier 1 ÷ an exposure including off-balance-sheet items, at least 3%, ignoring risk weights.
  4. The 3 pillars, stress tests and FRTB. Basel's Pillar 1 sets minimum capital, Pillar 2 adds bank-specific requirements after supervisory review, and Pillar 3 requires public risk disclosure.
  5. AT1, bail-in, MREL and the limits that protect depositors. AT1 bonds convert or are written down when CET1 falls below their trigger, often 5.125%.
  6. Which capital rule binds: risk weights or leverage. A bank needs the larger of risk-weighted assets times its risk-based ratio and total exposure times the 3% leverage ratio.

Accounting and the three statements

How any change moves profit, cash and the balance sheet, and why profit is not cash: the first technical block of every banking interview.

  1. One change through the 3 statements. Start on the income statement, carry net income to the cash flow statement, adjust it for non-cash items, and check that the balance sheet balances.
  2. Non-cash expenses and the tax they save. A deductible non-cash expense cuts net income by the expense × (1 − tax rate) and raises cash by the expense × the tax rate.
  3. Working capital: why profit is not cash. A rise in an operating asset, such as receivables or inventory, uses cash; a rise in an operating liability, such as payables, brings cash in.
  4. Building the cash flow statement. Operating cash flow is net income plus non-cash items, minus gains on assets sold, minus the increase in working capital.
  5. When revenue and costs are recognised. Revenue is recognised when the goods or service are delivered, not when cash arrives; a prepaid cost waits on the balance sheet.
  6. Capitalise or expense, EBITDA or EBIT. Capitalising a cost moves it from operating expenses into capex, so EBITDA is higher while the cash leaves either way.
  7. Deferred tax. Paying less tax now than the accounts show creates a deferred tax liability; paying more now creates a deferred tax asset.
  8. Dividends and buybacks in the accounts. Dividends and buybacks bypass the income statement: cash and equity fall by the amount paid, in financing activities, and neither is tax-deductible.
  9. IFRS 16: one lease, 3 statements. Under IFRS 16 a lease becomes a right-of-use asset and a lease liability, so rent leaves EBITDA and depreciation and interest replace it.

SQL and data analysis

Write queries that return the right rows: count joins before running them, handle NULL without losing data, filter groups with HAVING, rank with window functions and keep queries fast and safe.

  1. Joins and how many rows they return. An INNER JOIN returns one row per matching pair, and a LEFT JOIN adds one row with NULL columns for each unmatched left row.
  2. NULL: three-valued logic. Any comparison with NULL is unknown, so WHERE status = NULL returns nothing and IS NULL is the only test.
  3. GROUP BY, HAVING and the order of evaluation. A query runs logically FROM, WHERE, GROUP BY, HAVING, SELECT, ORDER BY, so WHERE filters rows before grouping and HAVING filters groups.
  4. Window functions: running totals, ranks and changes. A window function adds a value to each row without collapsing them: SUM(pnl) OVER (ORDER BY day) is a running total.
  5. Indexes, transactions and silent traps. An index speeds reads but slows writes, and a composite index on (desk, trade_date) serves desk or desk and date but not date alone.
  6. When the planner ignores your index. An index pays only for selective filters: each matching row costs about one random page read, while a full scan reads every page in sequence.

Audit and transaction services

What auditors test and why, how they set materiality and read risk, what an opinion says, and how a buyer's due diligence turns reported EBITDA and net debt into a price.

  1. Assertions and audit evidence. Auditors test occurrence by tracing records to evidence and completeness by tracing evidence to records, and check cut-off around year end.
  2. Materiality and the audit risk model. Materiality is the size of misstatement that would change a reader's decision, often about 5% of profit before tax.
  3. Fraud risks: journal entries, bill-and-hold, related parties. Auditors test manual journal entries near year end because management can override controls.
  4. Opinions, going concern and later events. A qualified opinion says the accounts are fair except for a specific matter, and every audit assesses going concern for at least 12 months.
  5. Normalised EBITDA. A buyer's due diligence restates EBITDA to what the business earns on a recurring basis at market costs.
  6. Net debt, debt-like items and the working capital peg. Due diligence turns the headline price into what the seller receives: debt-like items and a working capital shortfall reduce it.
  7. Why an add-back is worth the multiple and a debt-like item only once. An adjustment to EBITDA moves the price by the multiple, while a debt-like item or working capital shortfall moves it once, euro for euro.

Credit analysis

Read a borrower the way a lender does: leverage that holds up to adjustments, cover after real cash needs, liquidity, covenants and what each creditor recovers.

  1. Leverage and what moves it. Leverage is debt divided by EBITDA, so a fall in margin raises it with no new borrowing.
  2. Adjusted debt and adjusted EBITDA. Agencies add leases, unfunded pensions after tax and likely guarantees to debt, and remove recurring 'one-off' add-backs from EBITDA.
  3. Coverage and debt service. The most conservative interest cover deducts capex, (EBITDA − capex) ÷ interest, and floating-rate debt passes a rate rise straight into cover.
  4. Liquidity and maturities. Liquidity compares the next 12 months' sources, cash, undrawn committed facilities and free cash flow, with the uses, chiefly debt falling due.
  5. How ratings are set. A rating combines business risk with financial risk, so the same leverage earns a better rating in a steadier business.
  6. Covenants and protections. Bond covenants are usually incurrence tests, applied only when the company acts, such as borrowing more.
  7. Recovery and priority. In a default, secured lenders are paid from their collateral first and unsecured creditors share what is left.
  8. How lenders think. A lender's upside is capped at its interest and principal while its losses are not, so credit analysis starts from the downside.
  9. Holdco or opco: where the bond sits. Bonds issued by a holding company recover only from what is left of the subsidiary's equity once the subsidiary's creditors are paid.

Time series and econometrics

Work with AR and MA models, spot unit roots and spurious regressions, test cointegration for pairs trades, model volatility clustering and avoid the backtest traps.

  1. AR(1): mean, variance, forecasts and half-life. An AR(1) process x_t = c + φ x_{t−1} + ε_t with |φ| < 1 has a long-run mean of c ÷ (1 − φ).
  2. Reading autocorrelation: ACF, PACF and tests. An MA(q) process's autocorrelation function (ACF) cuts off after q lags, and an AR(p)'s partial autocorrelation function (PACF) after p lags.
  3. Unit roots and stationarity. When φ = 1 in an AR(1) the series has a unit root: a random walk whose shocks are permanent and whose variance grows with time.
  4. Cointegration and spurious regressions. 2 non-stationary series are cointegrated when a linear combination of them is stationary, which the Engle-Granger method tests on the regression's residual.
  5. Volatility clustering: GARCH and EWMA. Volatility clustering, large moves following large moves, is modelled on the variance with GARCH, whose long-run variance is ω ÷ (1 − α − β).
  6. Backtest pitfalls and forecast errors. Overlapping return windows make observations autocorrelated, so standard errors look too small unless corrected.
  7. How a GARCH forecast slopes back after a shock. A GARCH(1,1) variance forecast k days ahead is the long-run variance plus today's gap to it shrunk by (α + β) to the power k.

Cyber security

Think like attacker and defender: store passwords properly, tell hashing from encryption, close the classic web flaws, spot the scams aimed at banks and respond to an incident in order.

  1. Passwords, brute force and second factors. Passwords are stored with a slow, salted hash such as bcrypt or Argon2, never a fast hash like SHA-256.
  2. Hashes: salts, one-way functions and checksums. A per-user salt makes 2 identical passwords hash differently and defeats precomputed tables.
  3. Encryption, TLS, certificates and keys. TLS uses public-key cryptography only to agree on a shared key, then a fast symmetric cipher such as AES.
  4. Web flaws: injection, XSS, CSRF and access control. Parameterised queries send input as data so it cannot become SQL, and encoding output stops stored XSS.
  5. How attackers go after banks. A message from 'IT' with a look-alike link is reported and the real address typed by hand.
  6. Defence: privilege, zero trust, patching, detection and response. Each account gets only the rights it needs, and zero trust checks identity and device on every request wherever it comes from.
  7. How a slow hash turns hours of cracking into years. The time to try every password is the search space divided by the guesses a second the hash allows.

ESG and sustainable finance

Read sustainability claims like an analyst: SFDR fund articles, scopes 1, 2 and 3, carbon intensity and carbon costs, the taxonomy and double materiality, engagement and voting, and labelled bonds.

  1. SFDR: Articles 6, 8 and 9, adverse impacts and greenwashing. Under SFDR an Article 8 fund promotes environmental or social characteristics, an Article 9 fund targets sustainable investment, and overstating either is greenwashing.
  2. Scopes 1, 2 and 3, avoided emissions and attribution. Scope 1 covers a company's direct emissions, scope 2 the energy it buys and scope 3 the rest of its value chain.
  3. Carbon intensity, carbon costs, footprints and net zero claims. Carbon intensity is tonnes ÷ revenue in millions, and selling high emitters lowers a portfolio's footprint, not real emissions.
  4. The EU taxonomy, double materiality, nature and ratings. The EU taxonomy defines which activities count as environmentally sustainable, with the aligned share of revenue, capex and opex reported.
  5. Engagement, exclusion, voting and governance. Exclusion removes companies while engagement keeps them and uses dialogue and votes to push for change.
  6. Green, social and sustainability-linked debt. Green and social bonds fund eligible projects and report on them, while a sustainability-linked bond ties its coupon to company-wide ESG targets.
  7. A carbon price shock on equity value. A carbon price rise cuts EBITDA by the extra cost per tonne × emissions × the share not passed on.

IFRS accounting

Apply the IFRS rules analysts meet daily: revenue in 5 steps, leases, provisions and deferred tax, instruments and hedges, fair value and impairment, groups, segments and diluted EPS.

  1. Revenue under IFRS 15. IFRS 15 recognises revenue in 5 steps, contract, obligations, price, allocation, recognition, so a service paid upfront is spread over its term.
  2. Leases, provisions and deferred tax. IFRS 16 books leases as a right-of-use asset and a liability at the present value of payments, at the implicit or borrowing rate.
  3. Financial instruments and hedge accounting. Under IFRS 9 a debt instrument sits at amortised cost only if held to collect cash flows that are solely principal and interest.
  4. Fair value levels and impairment. IFRS 13 ranks fair values Level 1 for quoted prices, Level 2 for other observable inputs and Level 3 for unobservable ones.
  5. Groups: control, joint ventures, associates and currencies. Under IFRS 10 control means power, variable returns and the link between them, and joint ventures and associates use the equity method.
  6. Segments and diluted EPS. IFRS 8 defines operating segments the way the chief operating decision maker reviews the business, the management approach.
  7. Amortised cost: income at the effective rate. At amortised cost a bond's interest income is its carrying amount times the effective interest rate, the yield at purchase, not the coupon.

Say them out loud

The technical round of the video interview practice asks 5 questions on your desk, 90 seconds each, and reviews every answer. Each guide ends with a timed quiz.

Every topic comes from the Academy guides of this desk. Updated .