Probability
Recurrence Relations in Probability
Grade 12
Question:
<p><strong>For Problems 19–21:</strong> A player tosses a coin and scores one point for every head and two points for every tail that turns up. He plays on until his score reaches or passes \(n\). \(P_n\) denotes the probability of getting a score of exactly \(n\).</p><p>The value of \(P_n\) is equal to</p>
<p>(1) \((1/2)[P_{n-1} + P_{n-2}]\)</p>
<p>(2) \((1/2)[2P_{n-1} + P_{n-2}]\)</p>
<p>(3) \((1/2)[P_{n-1} + 2P_{n-2}]\)</p>
<p>(4) none of these</p>
Step-by-Step Solution
Key Concept: Set up a recurrence relation by considering the last toss: to reach exactly n, the player must have been at position (n-1) or (n-2) before the final toss, with probability 1/2 each. This gives P_n = (1/2)P_{n-1} + (1/2)P_{n-2}.
<p><strong>Step 1:</strong> Define base cases. The player starts at score 0, so P_0 = 1. With one toss: getting exactly 1 point requires a head (probability 1/2), so P_1 = 1/2. Getting exactly 2 points requires a tail (probability 1/2), so P_2 = 1/2.</p><p><strong>Step 2:</strong> For n ≥ 3, establish recurrence relation. To reach exactly score n, the player must either:</p><ul><li>Be at score (n-1) and toss a head (1 point): contributes (1/2)P_{n-1}</li><li>Be at score (n-2) and toss a tail (2 points): contributes (1/2)P_{n-2}</li></ul><p>Therefore: <strong>P_n = (1/2)P_{n-1} + (1/2)P_{n-2}</strong></p><p><strong>Step 3:</strong> Solve the recurrence. The characteristic equation is r² = (1/2)r + (1/2), giving 2r² - r - 1 = 0, so (2r + 1)(r - 1) = 0. Roots are r = 1 and r = -1/2.</p><p>General solution: P_n = A(1)ⁿ + B(-1/2)ⁿ = A + B(-1/2)ⁿ</p><p><strong>Step 4:</strong> Apply initial conditions. From P_0 = 1: A + B = 1. From P_1 = 1/2: A - B/2 = 1/2. Solving: A = 2/3, B = 1/3.</p><p>∴ <strong>P_n = 2/3 + (1/3)(-1/2)ⁿ</strong> or equivalently <strong>P_n = (2^{n+1} + (-1)^n)/(3·2ⁿ)</strong></p>
Correct Answer: A