Foundations of Risk Management
20%of exam
Quantitative Analysis
20%of exam
Financial Markets and Products
30%of exam
Valuation and Risk Models
30%of exam
Quick Facts
- Exam owner
- GARP
- Curriculum
- 2026 FRM Part I
- Questions
- 100 equally weighted
- Format
- Computer-based multiple choice
- Time limit
- 4 hours
- Average pace
- 2 min 24 sec per question
- Results
- Pass/fail with quartile feedback
- Exam months
- May, August, November
- Topic weights
- Approximate
Capacity, Appetite, Limits
Capacity can, appetite wants, limits bound
Systematic vs Idiosyncratic Risk
Systematic
- Common factor movements
- Market exposure measured by beta
Idiosyncratic
- Issuer-specific shocks
- Diversification reduces contribution
Diversification targets asset-specific risk
Performance Measure Picker
- Risk budget uses total volatility→Sharpe ratio(Compare risk-free excess return)
- Risk budget uses market beta→Treynor ratio(Systematic risk denominator)
- Compare return against CAPM→Jensen alpha(Subtract model-implied return)
- Compare skill against chosen benchmark→Information ratio(Scale by tracking error)
- Measure volatility of benchmark gap→Tracking error(Use active returns)
- Penalize shortfalls below target→Sortino ratio(Use downside deviation)
Risk and Governance
- Risk capacity
- Maximum risk absorbable
- Risk appetite
- Risk willingly assumed
- Risk limits
- Operational bounds on exposures
- Funding liquidity risk
- Difficulty meeting cash or collateral needs
- Market liquidity risk
- Difficulty exiting without price impact
- Board
- Oversees strategy and risk appetite
- Enterprise risk management
- Manage exposures across business units
- BCBS 239 data aggregation
- Accuracy, completeness, timeliness, adaptability
- Credit default swap buyer
- Pays premium for credit-event protection
- Equity tranche
- Absorbs pool losses first
- GARP conflicts
- Disclose fully to affected parties
- GARP confidentiality
- Safeguard information against improper disclosure
- GARP professional judgment
- Disclose relevant limits of expertise
Portfolio Risk and CAPM
- Expected portfolio return
- Σ(weight×expected return)
- Two-asset portfolio variance
- w₁²σ₁² + w₂²σ₂² + 2w₁w₂ρσ₁σ₂
- Portfolio notation
- w weight, σ volatility, ρ correlation
- Systematic risk
- Common market factor exposure
- Idiosyncratic risk
- Asset-specific diversifiable exposure
- Beta to market
- Cov(asset,market)/Var(market)
- CAPM expected return
- Rᶠ + β[E(Rₘ)−Rᶠ]Rᶠ risk-free rate
- Security market line
- Expected return against beta
- Capital market line
- Efficient portfolios by total volatility
- Arbitrage pricing theory
- Factor loadings times risk premiumsAdd risk-free rate
Performance Measures
- Sharpe ratio
- Risk-free excess return ÷ volatility
- Treynor ratio
- Risk-free excess return ÷ beta
- Jensen alpha
- Portfolio return minus CAPM return
- Tracking error
- Standard deviation of active return
- Information ratio
- Mean active return ÷ tracking error
- Active return
- Portfolio return minus benchmark return
- Sortino ratio
- Target excess return ÷ downside deviation
Testing Errors
One cries wolf, two misses wolf
Type I vs Type II Errors
Type I
- Reject a true null
- Probability α for exact-size test
Type II
- Miss a false null
- Probability β at given alternative
Power equals one minus β
Statistical Method Picker
- Update probability given evidence→Bayes rule(Divide by evidence probability)
- Normal mean, unknown variance→One-sample t test(Independent observations)
- Compare several coefficients jointly→F test(Use appropriate error assumptions)
- Regression errors have unequal variance→Heteroskedasticity-robust standard errors(Adjust inference)
- Remove a simple unit root→First differencing(Check differenced series)
- Estimate sampling variability empirically→Bootstrap(Choose dependence-aware resampling)
- Simulate from specified dynamics→Monte Carlo(Model drives scenarios)
Probability and Distributions
- Conditional probability P(A|B)
- P(A∩B)/P(B)P(B) > 0
- Bayes P(A|B)
- Prior×likelihood ÷ P(B)Likelihood is P(B|A)
- Total probability
- ΣP(B|Aᵢ)P(Aᵢ)Aᵢ partition outcomes
- Independent events
- P(A∩B)=P(A)P(B)
- Correlation ρ
- Cov(X,Y)/(σₓσᵧ)
- Bernoulli
- Single success-failure trial
- Binomial
- Success count in independent trialsFixed n and common p
- Poisson
- Event count at constant rateIndependent increments
- Student t
- Heavier tails than normal
- Lognormal X
- ln(X) is normalX strictly positive
Bootstrap vs Parametric Monte Carlo
Nonparametric bootstrap
- Resample observed data
- Tied to empirical sample
Parametric Monte Carlo
- Draw from specified distributions
- Depends on model assumptions
Identify source of each draw
Inference and Regression
- Unbiased sample variance
- Σ(xᵢ−x̄)²/(n−1)
- Mean standard error
- s/√n estimatedIndependent sample
- One-sample mean t
- (x̄−μ₀)/(s/√n)n−1 degrees of freedom
- Two-sided t interval
- x̄ ± critical t×s/√n
- Small p-value
- Reject H₀ when p ≤ α
- Test power
- 1 − Type II error probability
- Ordinary least squares
- Minimize squared residuals
- Regression coefficient
- Slope holding other regressors fixed
- Heteroskedasticity
- Nonconstant conditional error variance
- Multicollinearity
- Regressor overlap inflates standard errors
- Adjusted R²
- Penalizes added predictors
Time Series and Simulation
- Covariance stationarity
- Constant mean and finite varianceAutocovariance depends only on lag
- AR term
- Uses past observations
- MA term
- Uses past innovations
- AR(1) stationarity
- |φ| below 1φ lag coefficient
- Random walk
- Unit root in levelsFirst differences remove unit root
- Log return
- ln(Pₜ/Pₜ₋₁)Price return only
- Bootstrap
- Resample observations with replacement
- Monte Carlo
- Simulate from specified model
- Logistic regression
- Models log odds linearly
- Ridge
- L2 penalty on coefficients
- LASSO
- L1 penalty can zero coefficients
- Overfitting
- Weak generalization despite training fit
Buyer and Seller Hedges
Buyer buys futures, seller sells futures
Forward vs Futures
Typical forward
- Bilateral and customizable
- Usually settles at maturity
Futures
- Exchange-traded and standardized
- Marked to market daily
Check cash-flow timing and collateral
Hedge Direction Picker
- Future buyer fears price rises→Long futures hedge(Gains as futures rise)
- Future seller fears price falls→Short futures hedge(Gains as futures fall)
- Foreign-currency receivable→Sell foreign currency forward(Lock domestic proceeds)
- Foreign-currency payable→Buy foreign currency forward(Lock domestic cost)
- Floating-rate borrower fears rate rises→Pay fixed, receive floating(Match reset and notional)
- Stockholder wants downside floor→Buy a put(Premium buys protection)
- Hedge a long bond position→Short bond futures(Match DV01 per contract)
Institutions and Clearing
- Bank maturity transformation
- Short funding supports longer assets
- Defined benefit pension
- Sponsor bears investment risk
- Defined contribution pension
- Participant bears investment risk
- Open-end fund
- Redeems shares at net asset value
- Central counterparty
- Interposes between clearing counterparties
- Initial margin
- Collateral at position entry
- Maintenance margin
- Minimum required account equity
- Variation margin
- Settles current mark-to-market changes
European vs American Exercise
European
- Exercise only at expiry
- Parity prices matched call-put pairs
American
- Exercise through expiry
- Early exercise changes parity relationship
Exercise style drives pricing method
Forwards and Futures
- Basis convention
- Spot minus futures
- Basis risk
- Hedge instrument and exposure diverge
- Long forward payoff
- S_T − K at deliveryS_T delivery-date spot
- Fair forward with yield q
- S₀×exp[(r−q)T]r continuous risk-free rate
- Long forward value today
- D×(F−K)F current same-maturity forward price
- Discount factor D
- exp(−rT)r continuous rate, T years
- Commodity forward carry
- r + storage rate − convenience yieldContinuously compounded rates
- Commodity forward with carry k
- S₀×exp(kT)Constant net carry k
- Covered FX forward
- S×exp[(rᵈ−rᶠ)T]Domestic currency per foreign unit
- FX rate inputs
- rᵈ domestic, rᶠ foreignContinuous rates
- Minimum-variance hedge ratio
- ρ×σₛ/σᶠSpot/futures price changes
- Hedge contract count
- h×hedged units ÷ contract unitsPositive h shorts long spot
Rates and Swaps
- Fixing rate into future period
- Forward rate agreement
- Pay-fixed swap value
- PV floating leg minus fixed
- Par swap fixed rate
- Equates fixed and floating PVs
- Floating-rate debt hedge
- Pay fixed, receive floating
- Currency swap
- Cash flows in two currencies
- Futures delivery invoice
- Futures price×conversion factor + accrued interestSame face-value basis
- Cheapest to deliver
- Typically highest implied repo rateTreasury futures
- MBS contraction risk
- Falling rates accelerate prepaymentsTypical fixed-rate pool
- MBS extension risk
- Rising rates slow prepaymentsTypical fixed-rate pool
Option Positions
- Long call payoff
- max(S_T−K,0)
- Long put payoff
- max(K−S_T,0)
- Expiry payoff inputs
- S_T terminal spot, K strike
- European put-call parity
- C − P = S₀ − KDNo dividends. D discount factor.
- Known dividend adjustment
- Replace S₀ with S₀−PV(dividends)
- Protective put
- Long stock and long put
- Covered call
- Long stock and short call
- Long straddle
- Buy same-strike call and putSame expiry
- Asian option
- Payoff uses averaged underlying prices
- Barrier option
- Activation depends on barrier touch
Tail Threshold and Severity
VaR marks the cutoff, ES averages the tail
VaR vs Expected Shortfall
VaR
- Quantile threshold
- May fail subadditivity
Expected shortfall
- Worst-tail mean
- Subadditive for integrable losses
Use the same confidence and horizon
Valuation Method Picker
- Need threshold at chosen confidence→VaR(Fix horizon and loss sign)
- Need severity in worst tail→Expected shortfall(Average tail losses)
- Linear P&L with normal factors→Delta-normal VaR(Use factor covariance)
- Given joint historical factor changes→Historical simulation(Apply to current holdings)
- Need mean-reverting conditional variance→Stationary GARCH(1,1)(Persistence α+β below one)
- Small parallel bond yield shift→Modified duration(Add convexity for curvature)
- Yield curve twists→Key-rate duration(Shock points separately)
- Option permits early exercise→Binomial tree(Check exercise at each node)
Loss and Volatility Models
- Value at risk (VaR)
- Loss quantile at specified confidenceFix the horizon
- Expected shortfall (ES)
- Average of worst tail lossesWorst 1−confidence fraction
- Normal loss VaR
- μ + z×σμ loss mean, σ loss volatility
- Normal 95% and 99% z
- 1.645 and 2.326One-sided upper quantiles
- Square-root-of-time volatility
- σₕ = σ₁√hUncorrelated equal-variance increments
- Historical simulation
- Reprice using historical factor shocks
- VaR exception
- Realized loss exceeds forecast VaR
- EWMA next variance
- λv + (1−λ)ε²v prior variance, ε latest innovation
- GARCH(1,1) next variance
- ω + αε² + βvSame v and ε as EWMA
- GARCH long-run variance
- ω/(1−α−β)Stationary when α+β < 1
- Implied volatility
- Invert model using option price
Modified Duration vs DV01
Modified duration
- Relative sensitivity to yield
- Multiply by decimal yield change
DV01 magnitude
- Currency per basis point
- Scales with position value
Match units before sizing hedge
Credit and Stress
- Probability of default
- PD over specified horizon
- Loss given default
- LGD fraction of exposure lost
- Exposure at default
- EAD amount exposed at default
- Expected credit loss amount
- PD×LGD×EADBasic fixed-EAD model
- Loss given recovery R
- LGD = 1 − RR fraction recovered
- Constant hazard default probability
- 1 − exp(−λT)λ hazard, T horizon
- Point-in-time rating
- Responsive to current credit conditions
- Through-the-cycle rating
- Less sensitive to cyclical shifts
- Wrong-way risk
- Exposure rises with default risk
- Sovereign risk
- Government borrower credit risk
- Operational risk
- Process, people, system, external losses
- Sensitivity stress test
- Shock a selected risk factor
- Scenario stress test
- Revalue under coherent joint shocks
- Reverse stress test
- Find scenarios causing specified failure
Bond Valuation and Sensitivity
- Spot-curve bond value
- Σ(cash flow×date-specific discount factor)
- Yield to maturity
- Single rate equating price and PV
- Dirty bond price
- Clean price plus accrued interest
- Macaulay duration D_mac
- PV-weighted cash-flow time
- Modified duration
- D_mac/(1+y/m)y nominal yield, m periods/year
- Duration-convexity price change
- ΔP/P ≈ −D×Δy + ½C×Δy²D modified duration, C convexity
- Convexity C
- Relative price curvature versus yield
- DV01 magnitude
- Approx. |D×P|×0.0001P full position value
- Effective duration
- (P₋−P₊)/(2×P₀×Δy)Reprice with shifted curves
- Effective duration prices
- P₋ down, P₊ up, P₀ base
- Key-rate duration
- Sensitivity to selected curve maturity
Option Pricing
- Binomial up probability p
- (exp(rΔt)−d)/(u−d)No dividends, r continuous
- Binomial u and d
- Gross up and down multipliers
- Binomial continuation value
- Discount p-weighted next-node valuesDown weight 1−p
- American option node
- Max of exercise and continuation
- BSM log moneyness m
- ln(S₀/K)European, no dividends
- BSM d₁ rate term b
- r + σ²/2σ annualized volatility
- BSM d₁
- (m+bT)/(σ√T)
- BSM d₂
- d₁ − σ√T
- BSM call value
- S₀×N(d₁)−KD×N(d₂)N standard normal CDF
- BSM put value
- KD×N(−d₂)−S₀×N(−d₁)D risk-free discount factor
Option Greeks
- Delta
- Option value sensitivity to spot
- Gamma
- Delta sensitivity to spot
- Vega
- Option value sensitivity to volatility
- Theta
- Option value sensitivity to time
- Rho
- Option value sensitivity to rates
Common Traps
Dependence
Zero linear correlation ≠ Statistical independence
Hypothesis testing
Fail to reject H₀ ≠ Prove H₀ true
Forward terminology
Delivery price K fixed at inception ≠ Current fair forward price F
Option accounting
Payoff before premium ≠ Profit after premium adjustment
Futures margin
Performance collateral ≠ Option purchase premium
Interest rate units
1 basis point ≠ 1 percentage point
Tail risk
VaR quantile threshold ≠ Maximum possible loss
Option valuation probabilities
Risk-neutral pricing probabilities ≠ Real-world event probabilities
Credit loss inputs
Default probability ≠ Loss fraction conditional on default
Last Minute
- 1.Match returns and volatility horizons
- 2.Include covariance in portfolio variance
- 3.Scale Sharpe by total volatility
- 4.Divide Bayes by evidence probability
- 5.Check one-sided or two-sided tails
- 6.Match compounding and time units
- 7.Set futures direction from exposure
- 8.Discount strike in European parity
- 9.Include premium when calculating profit
- 10.Use decimal yields in duration
- 11.One basis point equals 0.0001
- 12.State VaR confidence and horizon
- 13.Expected shortfall averages worst losses
- 14.Test American exercise at each node
Explore More GARP FRM and Risk Certifications
Continue into nearby exams from the same family. Each card keeps practice questions, study guides, flashcards, videos, and articles in one place.
More From This Family
Videos and articles for deeper review.
