← Morgan Stanley Interview Insights

Morgan Stanley·Software Engineer·Technical Phone Screen·Senior

Senior
Apr 2026

Summary

Quant Engineer interview at Morgan Stanley that went deep into Black-Scholes delta-hedging theory. The whole session basically revolved around one big multi-part question on PnL attribution under misspecified vol, which sounds manageable until you're actually in it.

Questions Asked (5)

Q1

Walk through the structure of delta-hedging PnL when you hedge an option using one implied vol but the stock realizes a different vol over the option's life.

Technical Trade-offsAlgorithms & Data Structures
Author's notes

This is the kind of question where you know the answer exists and you've seen it before but reconstructing it under pressure is a different thing entirely.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by defining the setup: you hedge with implied vol σ_i but the stock realizes σ_r. Then decompose the PnL into the gamma-theta trade-off, showing that the daily PnL is approximately 0.5 * Γ * S^2 * (σ_r^2 - σ_i^2) * dt. Finally, integrate over the option's life to get the total PnL, which is path-dependent and proportional to the variance spread.

Pro tip: Emphasize that the PnL is not simply a function of the difference in vols but depends on the realized path of the stock price and the gamma profile. Mention that if realized vol exceeds implied vol, the hedger profits, and vice versa, but the exact PnL can vary due to path dependency.

1. Set up the hedging scenario

Define that you sell an option and hedge it using Black-Scholes with implied volatility σ_i, but the stock actually realizes volatility σ_r. Assume continuous hedging for simplicity.

2. Derive the daily PnL

Using Itô's lemma, show that the change in the hedged portfolio value over a small time step dt is approximately 0.5 * Γ * S^2 * (σ_r^2 - σ_i^2) * dt, where Γ is the option's gamma.

3. Integrate over the option's life

Sum the daily PnL from t=0 to T to get the total PnL: ∫ 0.5 * Γ_t * S_t^2 * (σ_r^2 - σ_i^2) dt. This shows the PnL is the integral of the gamma-weighted variance spread.

4. Interpret the result

Explain that if σ_r > σ_i, the hedger profits on average, and if σ_r < σ_i, the hedger loses. The magnitude depends on the path of S and the gamma profile.

5. Discuss practical implications

Mention that in reality, hedging is discrete, leading to additional hedging error, and that the PnL can be decomposed into a variance swap-like payoff plus a path-dependent residual.

Key Points to Mention

  • Black-Scholes delta hedging assumes continuous rebalancing and constant volatility.
  • The PnL from volatility mismatch is captured by the gamma-theta relationship: theta + 0.5 * Γ * S^2 * σ_i^2 = 0 under BS.
  • Daily PnL ≈ 0.5 * Γ * S^2 * (σ_r^2 - σ_i^2) * dt.
  • Total PnL is the integral of this expression over the option's life, which is path-dependent.
  • If realized vol > implied vol, the hedger profits; if realized vol < implied vol, the hedger loses.
  • Discrete hedging introduces additional noise and can lead to hedging errors beyond the volatility mismatch.

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.

Q2

Under what conditions does delta-hedging make or lose money, and how does that relate to gamma and the spread between realized and implied variance?

Technical Trade-offsPricing & Monetization
Author's notes

Easier to articulate once you've done the derivation above.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by explaining that delta-hedging isolates exposure to realized variance versus implied variance. Then describe how gamma determines the profit or loss from rebalancing, and conclude with the relationship between the P&L and the spread between realized and implied variance.

Pro tip: Emphasize that the sign of the P&L depends on whether realized variance exceeds implied variance, and that gamma is the key driver. Mention that in practice, transaction costs and discrete hedging can erode profits, so the theoretical result assumes continuous hedging and no costs.

1. Define Delta-Hedging and Its Goal

Explain that delta-hedging involves continuously adjusting the underlying position to keep the portfolio delta-neutral, aiming to eliminate directional risk.

2. Introduce Gamma and Its Role

Describe gamma as the rate of change of delta, which measures the curvature of the option price. Highlight that gamma determines the frequency and size of rebalancing trades.

3. Explain the P&L from Gamma Hedging

State that the profit or loss from delta-hedging is approximately half the sum of gamma times the squared change in the underlying price, i.e., 0.5 * Γ * (ΔS)^2.

4. Relate to Realized vs. Implied Variance

Connect the P&L to the difference between realized variance (from actual price moves) and implied variance (from option prices). If realized variance is higher than implied, a long gamma position profits; if lower, it loses.

5. Summarize Conditions for Profit/Loss

Conclude that delta-hedging makes money when realized variance exceeds implied variance (for long gamma) and loses when realized variance is lower, assuming no transaction costs.

Key Points to Mention

  • Delta-hedging aims to eliminate directional risk but leaves exposure to volatility.
  • Gamma measures the sensitivity of delta to changes in the underlying price.
  • The P&L from delta-hedging is approximately 0.5 * Γ * (ΔS)^2 per rebalancing period.
  • Realized variance is the actual variance of the underlying's returns over the hedging period.
  • Implied variance is the market's expectation of future variance, embedded in option prices.
  • Profit/loss depends on the spread: long gamma profits if realized > implied, loses if realized < implied.
  • Transaction costs and discrete hedging can reduce or eliminate theoretical profits.

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.

Q3

Sketch the result that total delta-hedging PnL approximates the integral of one-half times S-squared times Gamma times the difference in squared vols over time.

Algorithms & Data StructuresTechnical Trade-offs
Author's notes

I'd seen this formula before but never had to derive it from scratch on the spot.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by clarifying the assumptions (continuous hedging, no transaction costs, Black-Scholes framework) and then derive the PnL decomposition. Explain that the total PnL from delta-hedging a short option position equals the integral of (1/2) * S^2 * Gamma * (σ_realized^2 - σ_implied^2) dt, and then sketch how this approximates the given expression when the difference in squared vols is small. Use a simple diagram or equation to illustrate the relationship.

Pro tip: Emphasize that this is a fundamental result in quantitative finance: it shows that a delta-hedged option position is essentially a bet on realized volatility versus implied volatility. Mention that in practice, discrete hedging and transaction costs cause the actual PnL to deviate, but the integral approximation holds well under frequent rebalancing.

1. State the setup and assumptions

Assume a Black-Scholes world with continuous hedging, no transaction costs, and a short option position. Define the delta-hedged portfolio and the PnL over a small time interval.

2. Derive the instantaneous PnL

Using Itô's lemma, show that the change in portfolio value over dt is approximately (1/2) * S^2 * Gamma * (σ_realized^2 - σ_implied^2) dt, where σ_realized is the actual volatility and σ_implied is the volatility used for pricing.

3. Integrate over time

Sum the instantaneous PnL over the life of the option to get the total PnL as the integral of (1/2) * S^2 * Gamma * (σ_realized^2 - σ_implied^2) dt.

4. Relate to the given expression

Note that the difference in squared vols is (σ_realized^2 - σ_implied^2). If we denote this as Δσ^2, the integral becomes ∫ (1/2) S^2 Γ Δσ^2 dt, which matches the question's description.

5. Sketch the result

Draw a graph with time on the x-axis and the integrand on the y-axis, or simply write the integral expression and explain its meaning: the PnL is the weighted sum of the variance spread, with weights proportional to S^2 and Gamma.

Key Points to Mention

  • Delta-hedging eliminates directional risk, leaving exposure to volatility (gamma).
  • The PnL formula: Total PnL ≈ ∫ (1/2) S^2 Γ (σ_realized^2 - σ_implied^2) dt.
  • Gamma is largest near the strike and for short-dated options, so the PnL is concentrated there.
  • In practice, discrete hedging and transaction costs cause tracking error, but the integral approximation is the theoretical benchmark.
  • This result is used to understand volatility arbitrage and the PnL of option market making.
  • The sign of the PnL depends on whether realized vol exceeds implied vol (for a long gamma position).

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.

Q4

Even if realized vol exceeds implied vol on average, can the path of the stock still cause you to lose money on a delta-hedged position? Why?

Technical Trade-offsRoot Cause Analysis
Author's notes

This tripped me up more than it should have.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by affirming that yes, a delta-hedged position can lose money even if realized volatility exceeds implied volatility on average, because the P&L depends on the path of volatility, not just its average. Explain that the hedging error is driven by the difference between realized variance and implied variance, weighted by gamma, and that path dependence (e.g., large moves when gamma is low, or small moves when gamma is high) can lead to losses. Conclude by emphasizing that the average realized vol is not sufficient; the timing and magnitude of moves matter.

Pro tip: Mention that the loss can be quantified by the integral of gamma times the difference between realized variance and implied variance, and that discrete hedging and transaction costs further exacerbate the issue. This shows you understand both the theoretical and practical aspects.

1. Clarify the P&L of a delta-hedged option

Explain that the profit or loss from delta hedging is approximately the integral of (0.5 * gamma * S^2 * (realized variance - implied variance)) over the hedging period. This sets the foundation for why average realized vol alone is insufficient.

2. Introduce path dependence

Discuss how the hedging error depends on the path of the underlying and the gamma profile. For example, if the stock makes large moves when gamma is low (e.g., when the option is deep in or out of the money), the gains from realized vol may be small, while losses from implied vol can dominate.

3. Provide a concrete scenario

Illustrate with a scenario: Suppose you are short an at-the-money option. If the stock stays near the strike (high gamma) with low realized vol, you lose due to time decay. Then, if it suddenly jumps away (low gamma) with high realized vol, the gains from that jump are insufficient to offset the earlier losses. Thus, even if average realized vol > implied vol, you lose.

4. Address discrete hedging and other frictions

Mention that in practice, hedging is discrete, and transaction costs and slippage can cause additional losses. Also, the presence of jumps or stochastic volatility can make the hedging error path-dependent.

5. Conclude with the key insight

Summarize that the profitability of a delta-hedged position is not determined by average realized volatility alone, but by the covariance between gamma and realized variance. If the stock moves when gamma is low, the position can lose money even if realized vol exceeds implied vol on average.

Key Points to Mention

  • Delta-hedged P&L is approximately the integral of 0.5 * gamma * S^2 * (realized variance - implied variance) dt.
  • Path dependence: the timing of volatility relative to gamma matters.
  • Gamma is highest when the option is at-the-money and decreases as it moves in or out of the money.
  • If large moves occur when gamma is low, the gains from realized vol are small, and losses from implied vol can dominate.
  • Discrete hedging and transaction costs can cause additional losses.
  • The average realized volatility is not sufficient; the covariance between gamma and realized variance determines P&L.

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.

Q5

What are the practical implications of this analysis for choosing which vol to hedge at in real trading?

Technical Trade-offsPricing & Monetization
Author's notes

More open-ended and I actually preferred it.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by clarifying that the choice of hedging volatility depends on the specific product and risk profile, then discuss how the analysis informs practical decisions such as selecting the appropriate vol surface, managing model risk, and optimizing hedging costs. Emphasize the trade-offs between accuracy and computational efficiency, and how these impact real-time trading systems.

Pro tip: Demonstrate awareness that in practice, traders often hedge with a blend of implied vols from different sources (e.g., listed options, OTC quotes) and that the analysis should guide the weighting scheme rather than prescribe a single vol. Also, mention that regulatory constraints and capital requirements can influence the choice.

1. Identify the hedging objective

Clarify whether the goal is to minimize variance, match a specific benchmark, or reduce tail risk. This determines which vol (e.g., ATM, skew-adjusted, or realized) is most appropriate.

2. Assess available volatility measures

Compare implied vols from different sources (exchange-traded, broker quotes, model-implied) and realized vols. Consider liquidity, bid-ask spreads, and data quality.

3. Evaluate model and parameter risk

Discuss how the choice of vol interacts with the pricing model (e.g., Black-Scholes vs. stochastic vol). Highlight the sensitivity of hedges to vol misspecification and the need for robust calibration.

4. Consider operational constraints

Factor in transaction costs, margin requirements, and system latency. The analysis should inform whether a more complex vol surface is worth the computational overhead.

5. Implement and monitor

Propose a dynamic hedging strategy that adapts to changing market conditions, with regular backtesting and performance attribution to validate the vol choice.

Key Points to Mention

  • Volatility surface dynamics and the importance of skew and term structure
  • Trade-offs between model complexity and execution speed in real-time systems
  • Impact of transaction costs and liquidity on hedging effectiveness
  • Regulatory and capital considerations (e.g., Basel III, margin rules)
  • Backtesting and scenario analysis to validate vol selection
  • Communication with traders and risk managers to align on assumptions

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.