In: Statistics and Probability
Of all the cabs, 85 percent are blue and the rest are green. A cab identified in a hit-and-run accident is identified as green. People can correctly identify the color of a cab 80 percent of the time. When surveyed, most Stanford students thought there was an 80 percent chance the cab is actually green. Do you agree? Explain.
No the observation made by most of the Stanford students is incorrect. It is given that :
P(cab is blue) = 0.85
So P(cab is green) = 1- 0.85 = 0.15
Note that : P(A | B) = Probability of an event A given event B
P(people correctly identify the colour of a cab) = P(cab identified as green | it is actually green) = 0.8
We need to find P(it is actually green | cab identified as green)
Now by Bayes Theorem we have,
P(it is actually green | cab identified as green) = (P(cab identified as green | it is actually green) x P(cab is green)) / (P(cab identified as green))
Now by law of total probabilities,
P(cab identified as green) = P(cab identified as green | it is actually green) x P(cab is green) + P(cab identified as green | it is actually blue) x P(cab is blue) = 0.8 x 0.15 + 0.2 x 0.85 = 0.29
So,
P(it is actually green | cab identified as green) = (0.8 x 0.15) / 0.29 = 0.4137931034
So clearly what most Stanford students thought comes out to be wrong as if just does not depend on the probability of correctly identifying a colour but also the probability of a cab being green.
Thank you. Please rate positively if satisfied!