Study guide
17.1 Non-Parallel Term Structure Shifts
FRM Part I Financial Risk Manager Study Guide Exam
Track your progress as you read
Finished sections get a check mark.
1 Introduction & Exam Overview
1.1 Current Exam Facts 1.2 Study Strategy & Domain Weights
2 Foundations: Risk Building Blocks & Governance
2.1 Building Blocks of Risk Management 2.2 How Firms Manage Financial Risk 2.3 Governance of Risk Management 2.4 Credit Risk Transfer Mechanisms
3 Foundations: Portfolio Models, Data & ERM
3.1 MPT and the CAPM 3.2 APT and Multifactor Models 3.3 Risk Data Aggregation & Reporting 3.4 Enterprise Risk Management
4 Foundations: Disasters, Crisis & Ethics
4.1 Learning from Financial Disasters 4.2 Anatomy of the Great Financial Crisis 4.3 GARP Code of Conduct
5 Quantitative Analysis: Probability Foundations
5.1 Fundamentals of Probability 5.2 Random Variables 5.3 Common Univariate Distributions 5.4 Multivariate Random Variables
6 Quantitative Analysis: Inference & Regression
6.1 Sample Moments 6.2 Hypothesis Testing 6.3 Linear Regression 6.4 Multiple Regression
7 Quantitative Analysis: Diagnostics & Time Series
7.1 Regression Diagnostics 7.2 Stationary Time Series 7.3 Non-Stationary Time Series 7.4 Returns, Volatility & Correlation
8 Quantitative Analysis: Simulation & Machine Learning
8.1 Simulation and Bootstrapping 8.2 Machine-Learning Methods 8.3 Machine Learning and Prediction
9 Financial Markets: Institutions & Derivatives Intro
9.1 Banks 9.2 Insurance Companies and Pension Plans 9.3 Fund Management 9.4 Introduction to Derivatives
10 Financial Markets: Clearing, Futures & Hedging
10.1 Exchanges and OTC Markets 10.2 Central Clearing 10.3 Futures Markets 10.4 Using Futures for Hedging
11 Financial Markets: FX, Forwards, Commodities & Options
11.1 Foreign Exchange Markets 11.2 Pricing Financial Forwards and Futures 11.3 Commodity Forwards and Futures 11.4 Options Markets
12 Financial Markets: Option Properties, Strategies & Rates
12.1 Properties of Options 12.2 Trading Strategies 12.3 Exotic Options 12.4 Properties of Interest Rates
13 Financial Markets: Bonds, MBS, Rate Futures & Swaps
13.1 Corporate Bonds 13.2 Mortgages and Mortgage-Backed Securities 13.3 Interest Rate Futures 13.4 Swaps
14 Valuation & Risk Models: VaR, Volatility & Ratings
14.1 Measures of Financial Risk 14.2 Calculating and Applying VaR 14.3 Measuring and Monitoring Volatility 14.4 External and Internal Credit Ratings
15 Valuation & Risk Models: Credit, OpRisk & Stress
15.1 Country Risk 15.2 Measuring Credit Risk 15.3 Operational Risk 15.4 Stress Testing
16 Valuation & Risk Models: Fixed-Income Valuation
16.1 Pricing, Discounting & Arbitrage 16.2 Interest Rates 16.3 Bond Yields and Return Calculations 16.4 Duration, Convexity & DV01
17 Valuation & Risk Models: Term Structure, Options & Greeks
17.1 Non-Parallel Term Structure Shifts 17.2 Binomial Trees 17.3 The Black-Scholes-Merton Model 17.4 Option Sensitivity Measures: The Greeks
17.1 Non-Parallel Term Structure Shifts
Key Takeaways Explain This Topic AI
PCA decomposes yield-curve moves into a few orthogonal factors—typically level, slope, and curvature—that explain most historical variance.
A key-rate duration (KR01) is the portfolio’s dollar (or DV01-style) sensitivity to a localized bump at one maturity while neighboring rates are interpolated to hold other key rates fixed.
Portfolio KR01s sum exposures across positions; hedging matches each key-rate bucket with instruments (bonds, futures, swaps) that concentrate risk in that bucket.
Forward-bucket 01 shocks contiguous forward rates; KR01 shocks par/spot key rates with tent-shaped interpolations—both recover effective duration when all buckets move in parallel.
PCA factor volatilities and key-rate volatilities drive multi-factor VaR and stress P&L; parallel-duration alone misses twist and butterfly risk.
Last updated: August 2026
Non-Parallel Term Structure Shifts
Parallel-shift duration (modified duration or DV01) answers only one question: what happens if every yield on the curve moves by the same amount. Real curves twist and butterfly . VRM–13 gives FRM candidates the two standard toolkits for non-parallel risk: principal component analysis (PCA) of historical curve moves, and key-rate / forward-bucket 01s that localize sensitivity along the maturity axis.
Why Parallel Shifts Are Not Enough
A barbell (cash in 2-year and 30-year bonds) and a bullet (cash in a 10-year bond) can be duration-matched so a +1 bp parallel move produces nearly identical P&L. A steepener—short rates up, long rates down—hurts the barbell and helps the bullet (or the reverse, depending on signs). Regulatory and desk stress tests therefore require shape risk , not only level risk.
Shift type Description Duration alone?
- Parallel (level) All yields +Δy Captured
- Steepener / flattener (slope) Short and long move oppositely Missed
- Butterfly (curvature) Belly moves vs wings Missed
- Localized spike One sector jumps (e.g., 5y) Missed
- PCA Drivers of Curve Moves
Take a vector of daily (or weekly) yield changes at fixed tenors, e.g., Δy = (Δy_2y, Δy_5y, Δy_10y, Δy_30y, …). Form the covariance (or correlation) matrix Σ of these changes. PCA finds orthonormal eigenvectors (factor loadings) and eigenvalues (factor variances):
Σ v_i = λ_i v_i
Order λ_1 ≥ λ_2 ≥ … The first principal component almost always looks like a roughly flat vector across tenors—the level factor. The second typically changes sign from short to long tenors—the slope factor. The third is often U-shaped or inverted-U—the curvature (butterfly) factor. In many government-bond markets, the first three PCs explain 90%+ of historical variance; higher PCs are residual noise or liquidity idiosyncrasies.
Worked PCA intuition
Suppose four tenors and estimated factor volatilities (annualized, in yield bp):
- Factor Interpretation σ (bp/year) Share of variance
- PC1 Level 80 82%
- PC2 Slope 25 12%
- PC3 Curvature 12 4%
- Residual Idiosyncratic — 2%
A portfolio’s PCA risk is the vector of factor exposures (dollar sensitivity to each PC shock) times these volatilities. Multi-factor VaR ≈ z_α × √(Σ_i (e_i σ_i)²) if factors are orthogonal (as PCs are by construction). Exam cue: orthogonality is why PCA is preferred to ad-hoc level/slope/curve regressions that leave correlated residuals.
Key-Rate Shifts and KR01
Key-rate duration (Ho’s framework, popularized in risk systems) defines a set of key maturities —commonly 6m, 2y, 5y, 10y, 20y, 30y. A key-rate shift at key maturity k raises that par (or zero) yield by 1 bp and interpolates neighboring points so that other key rates stay fixed . The interpolation is usually piecewise linear (“tent”) : the shift is 1 at key k, 0 at adjacent keys, and linear in between. The portfolio’s KR01_k (or key-rate DV01) is the dollar P&L (or change in value) from that localized +1 bp tent, often reported as −ΔV per bp so that long bonds have positive KR01 like classic DV01.
- Formally, if V(y) is portfolio value as a function of the curve,
- KR01_k ≈ −[V(y + tent_k) − V(y)] / 1 bp
- (Sign conventions vary; state yours clearly on the exam.)
- Worked KR01 for a single bond
A 10-year par bond with DV01 ≈ $850 per $1 million face under a parallel 1 bp shift. Under a pure 10y key-rate tent, almost the entire curve segment affecting that bond moves, so KR01_10y ≈ $820–$850 (slightly less if coupons create sensitivity to shorter keys). Under a pure 2y key-rate tent, the 10y par yield barely moves (tent is zero at 10y), so KR01_2y is small but not zero because the discount factors for near coupons still shift slightly depending on how the interpolator builds the zero curve from par keys.
- Portfolio KR01s
- For a book of positions i = 1…N,
- KR01_k(portfolio) = Σ_i KR01_k(i)
KR01s are additive across positions (to first order), which makes them ideal for limits and hedging grids. A portfolio plot of KR01 versus key maturity is the risk profile : peaks show where curve risk concentrates.
Worked portfolio aggregation
- Position KR01_2y KR01_5y KR01_10y KR01_30y
- Long 5y note 12 420 35 2
- Short 10y future −5 −40 −380 −15
- Receive-fixed 30y swap 8 50 90 510
- Net 15 430 −255 497
Units: dollars per bp. The book is long the 5y and 30y sectors and short the 10y sector—classic barbell versus belly risk even if parallel DV01 = 15 + 430 − 255 + 497 = 687 looks like a simple long.
Hedging Instruments for Key Rates
To flatten KR01_k, trade an instrument whose KR01 is concentrated at k:
- Bucket Typical hedge
- 2y 2y Treasury / CTZ, 2y futures, 2y swap
- 5y 5y note / FV futures, 5y swap
- 10y 10y note / TY futures, 10y swap
- 30y Long bond / US futures, 30y swap / bond future
Hedge ratio for bucket k: N_hedge ≈ −KR01_k(portfolio) / KR01_k(hedge per contract or per $ face). After hedging all keys, residual risk is interpolation error, convexity, spread (swap vs Treasury), and PCA factors beyond the key grid.
Worked hedge
Net KR01_10y = −$255 / bp. A 10y futures CTD basket has KR01 ≈ $85 / bp per contract. Number of contracts to buy ≈ 255 / 85 ≈ 3.0 contracts (buy because portfolio KR01_10y is negative—you need positive 10y duration). Recompute all buckets after the trade: 10y KR01 near zero, small leakage into 5y and 30y from the tent overlap.
PCA Volatility Versus Key-Rate Volatility
PCA volatilities σ_i = √λ_i are factor-level. Key-rate volatilities are the historical (or implied) volatilities of each key-rate change, plus a correlation matrix among keys. Both feed multi-factor VaR:
PCA route: map portfolio to PC exposures e_i, use diagonal factor vols.
Key-rate route: map to KR01 vector d, use dᵀ Σ_key d for variance of P&L (P&L ≈ −d · Δy_keys in dollar-bp units).
Key-rate vols are often higher at the short end (monetary-policy noise) and lower in the long end , with high correlations between adjacent keys. PCA compresses that correlated structure into a few orthogonal shocks—cleaner for stress narratives (“+2σ level, −1σ slope”).
Forward-Bucket 01 Versus KR01
A forward-bucket 01 (or forward DV01) shocks a contiguous segment of the instantaneous or discrete forward curve by 1 bp while leaving other forward segments unchanged. That is a rectangular shock in forward space, which translates into a smooth but different shape in par/zero space than a key-rate tent.
- Feature KR01 (key-rate) Forward-bucket 01
- Shock object Par or zero key rates Forward rates in a bucket
- Shape in zero space Tent / piecewise linear Hump from integrating forwards
- Natural hedges Par bonds, swaps, futures Forward-starting swaps, FRAs, Eurodollars
- Additivity Yes across keys Yes across buckets
- Parallel sum Sum of KR01s ≈ parallel DV01 Sum of bucket 01s ≈ parallel DV01
Exam distinction: if the vignette shocks “the 2y–5y forward,” think bucket 01 . If it bumps “the 5y par yield holding other key pars fixed,” think KR01 .
Duration from KR01 and Bucket 01
If every key rate (or every forward bucket) is shifted by the same 1 bp, the total P&L is the sum of the localized P&Ls. Therefore:
Parallel DV01 ≈ Σ_k KR01_k ≈ Σ_b Bucket01_b
Modified duration follows as D_mod ≈ (Σ KR01) / (0.0001 × V) when KR01 is dollar sensitivity per bp and V is value. This is a crucial consistency check: if your key-rate profile does not sum to the parallel DV01 within interpolation tolerance, the risk system is mis-specified.
Worked consistency check
Take a long-only book whose KR01s are ten times the net profile above: Σ KR01 = 6,870 $/bp and market value V = $10,000,000. Then
D_mod = (Σ KR01) / (V × 0.0001) = 6,870 / (10,000,000 × 0.0001) = 6,870 / 1,000 = 6.87 years .
If hedges cut the same shape down to Σ KR01 = 687 $/bp on the same V, parallel duration falls to 0.687 years —the key-rate shape can still be large even when net parallel duration looks tiny. Always reconcile Σ KR01 (or Σ bucket 01) with the system's parallel DV01 before trusting either number.
Synthesis for Exam Vignettes
When a question mentions “non-parallel,” reach for PCA level/slope/curvature or a key-rate grid—not a single duration number. Hedge each material KR01 bucket; verify that the sum of KR01s (or bucket 01s) recovers parallel DV01; and remember that PCA orthogonality simplifies multi-factor VaR while key-rate profiles communicate where on the curve the risk lives.