Detailed Explanation
This is a classic Bayes' theorem problem. We are given the probability of an accident for each category of driver, and we want to reverse the reasoning: given that an accident happened, find the probability that the driver was from a specific category (a scooter driver). First we choose an insured person at random, so the chance of picking each category is proportional to how many people are in it. These are the prior probabilities. The accident probabilities are conditional probabilities: the chance of an accident given the category. Bayes' theorem combines these to give the posterior probability, that is, the probability of the category given that an accident has occurred.
Formula Used
Let \( E_1, E_2, E_3 \) be the events that the person is a scooter, car and truck driver, and \( A \) the event of an accident. Then \[ P(E_1 \mid A) = \frac{P(E_1)\,P(A \mid E_1)}{P(E_1)P(A \mid E_1) + P(E_2)P(A \mid E_2) + P(E_3)P(A \mid E_3)} \]
Solution
- Step 1: Find the total number of insured drivers. Total \( = 2000 + 4000 + 6000 = 12000 \). This total is needed to compute the prior probability of picking a driver from each category.
- Step 2: Compute the prior probabilities (chance of selecting each type of driver). \[ P(E_1) = \frac{2000}{12000} = \frac{1}{6} \] \[ P(E_2) = \frac{4000}{12000} = \frac{1}{3} \] \[ P(E_3) = \frac{6000}{12000} = \frac{1}{2} \]
- Step 3: Write the conditional probabilities of an accident given each category, as provided in the question. \[ P(A \mid E_1) = 0.01, \quad P(A \mid E_2) = 0.03, \quad P(A \mid E_3) = 0.15 \]
- Step 4: Compute each product \( P(E_i)P(A \mid E_i) \), which represents the joint chance of that category and an accident. \[ P(E_1)P(A \mid E_1) = \frac{1}{6} \times 0.01 = \frac{0.01}{6} \] \[ P(E_2)P(A \mid E_2) = \frac{1}{3} \times 0.03 = \frac{0.03}{3} = \frac{0.06}{6} \] \[ P(E_3)P(A \mid E_3) = \frac{1}{2} \times 0.15 = \frac{0.15}{2} = \frac{0.45}{6} \]
- Step 5: Add these products to get the total probability of an accident (denominator of Bayes' theorem). \[ P(A) = \frac{0.01}{6} + \frac{0.06}{6} + \frac{0.45}{6} = \frac{0.52}{6} \]
- Step 6: Apply Bayes' theorem to find the required posterior probability. \[ P(E_1 \mid A) = \frac{\dfrac{0.01}{6}}{\dfrac{0.52}{6}} = \frac{0.01}{0.52} = \frac{1}{52} \]
- Step 7: Interpret the result: out of all insured persons who meet with an accident, the fraction who are scooter drivers is \( \frac{1}{52} \approx 0.0192 \).
Final Answer
The probability that the person who met with an accident is a scooter driver is \( \dfrac{1}{52} \approx 0.0192 \).
Common Mistake
A common mistake is to treat the prior probabilities as equal (like \( \frac{1}{3} \) each) instead of weighting them by the number of drivers in each category. Since there are different numbers of scooter, car and truck drivers, you must use \( P(E_1) = \frac{1}{6} \), \( P(E_2) = \frac{1}{3} \), \( P(E_3) = \frac{1}{2} \).