Page 222 #29
Suppose that the time (in hours) required to repair a car is an exponentially distributed random variable with parameter λ = 1/2.
The probability that the repair time exceeds 4 hours is equal to 1 minus the probability that it last 4 hours or fewer. With R:
1-pexp(4,.5)
## [1] 0.1353353
By hand: \[ 1-F(4) = 1-P(T ≤ 4)\\ = 1-\int_{0}^{4}0.5e^{−0.5t} \, dt\\ =1- (1 − e^{(-0.5)(4)})\\ =\frac {1}{e^2}\\ \approx 0.1353\\ \]
Since the exponential distribution has the memoryless property, if the time has already exceeded a certain amount, we “start over” at t=0. So the probability that it exceeds 8 hours, given that it has exceeded 4 hours, is the same as the probability that it exceeds 4 hours.
1-pexp(8-4,.5)
## [1] 0.1353353