Methods and Models — Valuation and Hedging Using Binomial Trees

Methods and Models — Valuation and Hedging Using Binomial Trees

CAIA Level II — High-Density Study Guide

What Is It?

1. What is this topic?
   A discrete-time model for pricing derivatives and hedging risk by simulating asset price movements over time using a tree of possible outcomes.
2. How is it tested, applied, or used?
   Tested via numerical problems (e.g., pricing options, calculating hedge ratios). Applied in real-world derivative pricing, risk management, and arbitrage strategies.

Why Does the Exam Ask This?

Tests quantitative reasoning, risk-neutral valuation, and dynamic hedging—core skills for alternative investments. Measures ability to: - Model uncertainty in discrete steps. - Apply no-arbitrage principles. - Construct replicating portfolios for hedging. - Interpret model outputs for trading or compliance.

What Do I Need to Know First?

1. Risk-neutral valuation (pricing under a risk-neutral measure).
2. No-arbitrage principle (law of one price).
3. Basic option pricing (payoffs, intrinsic value).
4. Probability trees (up/down movements, probabilities).
5. Replicating portfolio (delta hedging).

Topic Snapshot

Binomial trees are a foundational tool in CAIA Level II for pricing derivatives (e.g., options, swaps) and hedging in illiquid or complex markets. Unlike Black-Scholes (continuous-time), binomial trees handle discrete time steps, making them practical for real-world applications (e.g., private equity, real assets). Mastery is critical for structured products, exotic options, and risk management.

Exam / Job / Audit Weighting

• Frequency: High (appears in 1–2 questions per exam).
• Difficulty Rating: Intermediate.
• Question Type: Numerical calculations, conceptual applications, scenario-based hedging.

Difficulty Level

Intermediate (requires comfort with probability, algebra, and financial intuition).

Must-Know Rules, Formulas, Standards, or Principles

1. Risk-neutral probability (q):
   [
   q = \frac{(1 + r) - d}{u - d}
   ]
   where (u) = up-factor, (d) = down-factor, (r) = risk-free rate.

2. Option pricing via backward induction:

3. Start at expiration (payoff = intrinsic value).

4. Work backward, discounting expected payoffs at each node:
   [
   C = \frac{q \cdot C_u + (1-q) \cdot C_d}{1 + r}
   ]

5. Delta hedging (replicating portfolio):

6. Delta ((\Delta)) = number of shares to hold per option:
   [
   \Delta = \frac{C_u - C_d}{S_u - S_d}
   ]
7. Hedge by holding (\Delta) shares and borrowing/lending cash.

Misconceptions

1. "Binomial trees are just a simplified Black-Scholes."
   → They’re more flexible (handle dividends, American options, path-dependency).
2. "Risk-neutral probabilities are real-world probabilities."
   → They’re artificial (derived from no-arbitrage, not market expectations).
3. "More steps = more accuracy."
   → True, but computationally intensive; trade-off with speed.
4. "Delta hedging eliminates all risk."
   → Only instantaneous risk; requires dynamic rebalancing.
5. "Binomial trees can’t price American options."
   → They excel at early exercise (unlike Black-Scholes).

Common Mistakes

1. Incorrect risk-neutral probability:
2. Using (u) and (d) in the wrong order (must be (u > 1 + r > d)).
3. Forgetting to discount at each step:
4. Applying a single discount at the end (must discount per period).
5. Miscounting time steps:
6. Off-by-one errors (e.g., 3 steps = 4 nodes at expiration).
7. Ignoring dividends:
8. Not adjusting the stock price for dividends (reduces (S) at ex-date).
9. Misapplying delta hedging:
10. Holding the wrong number of shares (e.g., using (\Delta) from the wrong node).

The Common Trap

Assuming the tree is symmetric. - Many learners default to (u = 1/d) (e.g., (u = 1.2), (d = 0.833)), but real-world trees are often asymmetric (e.g., (u = 1.3), (d = 0.7)). - Trap: Using symmetric trees when the question implies volatility skew or non-standard movements.

Terms to Remember

1. Node: A point in the tree representing a possible asset price at a given time.
2. Up-factor (u): Multiplier for the asset price moving up (e.g., (u = 1.1) → +10%).
3. Down-factor (d): Multiplier for the asset price moving down (e.g., (d = 0.9) → -10%).
4. Replicating portfolio: A combination of stock and cash that mimics the option’s payoff.
5. Early exercise premium: Additional value in American options from the right to exercise early.

Step-by-Step Process

1. Build the Tree

• Inputs: Current stock price ((S_0)), up/down factors ((u, d)), risk-free rate ((r)), time steps ((n)).
• Steps:
• Calculate (S_u = S_0 \cdot u), (S_d = S_0 \cdot d).
• Repeat for each step (e.g., (S_{uu} = S_u \cdot u), (S_{ud} = S_u \cdot d)).
• Ensure (u > 1 + r > d) (no-arbitrage condition).

2. Calculate Risk-Neutral Probability (q)

• Use: [ q = \frac{(1 + r) - d}{u - d} ]
• Verify (0 < q < 1) (valid probability).

3. Price the Option (Backward Induction)

• At expiration (final nodes):
• Call: (C = \max(S - K, 0))
• Put: (P = \max(K - S, 0))
• Prior nodes:
• Discount expected payoffs:
  [
  C = \frac{q \cdot C_u + (1-q) \cdot C_d}{1 + r}
  ]
• For American options, compare with intrinsic value at each node.

4. Delta Hedging (Optional)

• Calculate (\Delta) at each node: [ \Delta = \frac{C_u - C_d}{S_u - S_d} ]
• Construct replicating portfolio:
• Hold (\Delta) shares.
• Borrow/lend cash to match option value.

Exam Answer Builder

1-Mark Question (Conceptual)

What it tests: Understanding of risk-neutral probability. Example: "In a binomial tree, the risk-neutral probability (q) is derived from:" A) Historical stock returns B) No-arbitrage conditions C) Investor risk preferences D) Market supply and demand Key Tip: Answer is B (no-arbitrage). Eliminate A (real-world), C (subjective), D (irrelevant).

2-Mark Question (Calculation)

What it tests: Basic option pricing. Example: "A stock trades at $50. In one period, it can move to $60 (u=1.2) or $40 (d=0.8). The risk-free rate is 5%. What is the price of a 1-period European call option with strike $50?" Key Tip: 1. Calculate (q = \frac{1.05 - 0.8}{1.2 - 0.8} = 0.625). 2. Payoffs: (C_u = \max(60 - 50, 0) = 10), (C_d = 0). 3. Price: (\frac{0.625 \cdot 10 + 0.375 \cdot 0}{1.05} = \$5.95).

5-Mark Question (Multi-Step)

What it tests: Full binomial tree pricing. Example: "A stock is at $100. Over two periods, it can move up 20% or down 10% each period. The risk-free rate is 3%. Price a 2-period European put option with strike $100." Key Tip: 1. Build tree: (S_{uu} = 144), (S_{ud} = S_{du} = 108), (S_{dd} = 81). 2. Final payoffs: (P_{uu} = 0), (P_{ud} = 0), (P_{dd} = 19). 3. Backward induction:
- (P_u = \frac{0.55 \cdot 0 + 0.45 \cdot 0}{1.03} = 0)
- (P_d = \frac{0.55 \cdot 0 + 0.45 \cdot 19}{1.03} = 8.25)
- (P_0 = \frac{0.55 \cdot 0 + 0.45 \cdot 8.25}{1.03} = \$3.60).

Case Study (Hedging Application)

What it tests: Dynamic delta hedging. Example: "You sell a 1-period European call option on a stock (current price $50, u=1.2, d=0.8, r=5%). The option’s delta is 0.625. How do you hedge the position?" Key Tip: 1. Hold 0.625 shares (long). 2. Borrow cash to finance the hedge:
- Cost of shares: (0.625 \cdot 50 = \$31.25).
- Option premium received: (C_0 = \$5.95) (from earlier).
- Borrow: (31.25 - 5.95 = \$25.30). 3. At expiration:
- If (S_u = 60): Sell shares for (0.625 \cdot 60 = \$37.50), repay loan ((25.30 \cdot 1.05 = \$26.57)), net profit = (37.50 - 26.57 = \$10.93) (matches (C_u = 10)).
- If (S_d = 40): Sell shares for (0.625 \cdot 40 = \$25), repay loan, net profit = (25 - 26.57 = -\$1.57) (matches (C_d = 0)).

This vs That

Time-Saver Hack

Recognize the "1-step shortcut" for European options: If the option is European and the tree has only 1 step, you can price it as: [ C = \frac{q \cdot C_u + (1-q) \cdot C_d}{1 + r} ] Skip building the full tree if the question is simple.

Mini Scenarios

1. Basic Scenario

"A stock is at $40. In one period, it can go to $50 or $30. The risk-free rate is 2%. What is the risk-neutral probability?" What to notice: - (u = 1.25), (d = 0.75). - Check no-arbitrage: (1.25 > 1.02 > 0.75) (valid). - Calculate (q = \frac{1.02 - 0.75}{1.25 - 0.75} = 0.54).

2. Applied Scenario

"You’re pricing a 2-period American put option. At the first down node, the option’s value is $5 if held, but the intrinsic value is $8. What do you do?" What to notice: - Early exercise is optimal (American option). - Replace the held value ($5) with intrinsic value ($8) at that node.

3. Tricky Scenario

"A binomial tree has (u = 1.1), (d = 0.9), and (r = 5\%). The risk-neutral probability (q) is 0.75. Is this tree arbitrage-free?" What to notice: - Check (u > 1 + r > d): (1.1 > 1.05 > 0.9) (valid). - But (q = 0.75) implies (1 + r = q \cdot u + (1-q) \cdot d): (1.05 = 0.75 \cdot 1.1 + 0.25 \cdot 0.9 = 1.05) (valid). - No arbitrage, but the high (q) suggests high volatility skew.

Diagnostic MCQ Bank

Easy

Question 1: "In a binomial tree, the up-factor (u) is 1.2 and the down-factor (d) is 0.8. If the risk-free rate is 5%, what is the risk-neutral probability (q)?" A) 0.50 B) 0.625 C) 0.75 D) 0.80 Correct Answer: B) 0.625 Explanation: [ q = \frac{1.05 - 0.8}{1.2 - 0.8} = 0.625 ] Trap Option: A) 0.50 (assumes symmetry, but (u \neq 1/d)).

Question 2: "A 1-period European call option has strike $50. The stock can move to $60 or $40. The risk-neutral probability is 0.6. What is the option’s value if the risk-free rate is 0%?" A) $6 B) $8 C) $10 D) $12 Correct Answer: A) $6 Explanation: [ C = \frac{0.6 \cdot 10 + 0.4 \cdot 0}{1 + 0} = 6 ] Trap Option: C) $10 (ignores probability weighting).

Medium

Question 3: "A stock is at $100. Over two periods, it can move up 10% or down 5% each period. The risk-free rate is 2%. What is the price of a 2-period European call option with strike $100?" A) $8.33 B) $9.45 C) $10.20 D) $11.00 Correct Answer: B) $9.45 Explanation: 1. Build tree: (S_{uu} = 121), (S_{ud} = 104.5), (S_{dd} = 90.25). 2. Final payoffs: (C_{uu} = 21), (C_{ud} = 4.5), (C_{dd} = 0). 3. (q = \frac{1.02 - 0.95}{1.10 - 0.95} = 0.4667). 4. Backward induction:
- (C_u = \frac{0.4667 \cdot 21 + 0.5333 \cdot 4.5}{1.02} = 12.35)
- (C_d = \frac{0.4667 \cdot 4.5 + 0.5333 \cdot 0}{1.02} = 2.06)
- (C_0 = \frac{0.4667 \cdot 12.35 + 0