Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Chapter 5. Continuous Probability Distributions Sections 5.4, 5.5: Exponential and Gamma Distributions Jiaping Wang Department of Mathematical Science 03/25/2013, Monday The UNIVERSITY of NORTH CAROLINA at CHAPEL HILL Outline Exponential: PDF and CDF Exponential: Mean and Variance Gamma: PDF and CDF Gamma: Mean and Variance More Examples The UNIVERSITY of NORTH CAROLINA at CHAPEL HILL Part 1. Exponential: PDF and CDF The UNIVERSITY of NORTH CAROLINA at CHAPEL HILL Probability Density Function In general, the exponential density function is given by π π₯ = 1 β π₯ /π π , π π₯β₯0 0, ππ‘βπππ€ππ π Where the parameter ΞΈ is a constant (ΞΈ>0) that determines the rate at which the curve decreases. ΞΈ=2 ΞΈ = 1/2 The UNIVERSITY of NORTH CAROLINA at CHAPEL HILL Cumulative Distribution Function The exponential CDF is given as 0, π₯ < 0 / πΉ π₯ = 1 β π β π₯ π, π₯ β₯ 0 ΞΈ=2 ΞΈ = 1/2 The UNIVERSITY of NORTH CAROLINA at CHAPEL HILL Part 2. Mean and Variance The UNIVERSITY of NORTH CAROLINA at CHAPEL HILL Gamma Function The gamma function Ξ(Ξ±) is given as β Ξ πΌ = 0 π₯πΌ β 1π β π₯ππ₯ We can show that Ξ πΌ + 1 = πΌΞ πΌ So Ξ π = π β 1 Ξ π β 2 = β― = π β 1 ! Specially, Ξ 1/2 = π The UNIVERSITY of NORTH CAROLINA at CHAPEL HILL Mean and Variance β π¬ πΏ = β ππ π π π = ββ = π ππ = π β π π½ π β πππ β π±π ππ±π© π π π± β π π β π½ π π π = ππ± = π πͺ π π π πππ β π π π½ π½ π Ξ π½ π π½π = π½. π ππ = πππ. Then we have V(X)=E(X2)-E2(X)=2ΞΈ2- ΞΈ2= ΞΈ2. The UNIVERSITY of NORTH CAROLINA at CHAPEL HILL Example 5.9 A sugar refinery has three processing plants, all of which receive raw sugar in bulk. The amount of sugar that one plant can process in one day can be modeled as having an exponential distribution with a mean of 4 tons for each of the three plants. If the plants operate independently, find the probability that exactly two of the three plants will process more than 4 tons on a given day. Answer: The probability that any given plant will process more than 4 tons a day, with X representing the amount used, is 4 π = π π > 4 = 1 β π π β€ 4 = 1 β πΉ 4 = 1 β 1 β exp β 4 = exp β1 = 0.37 As the plants operate independently, the problem is to find the probability of two successes out of three tries with p=0.37, which is a binomial distribution, so P(Exactly two of three plants use more than 4 tons)= 32 0.37 2 0.63 = 0.26. The UNIVERSITY of NORTH CAROLINA at CHAPEL HILL Example 5.10 Consider a particular plant in Example 5.9. How much raw sugar should be stocked for that plant each day so that the chance of running out of product is only 0.05? Answer: Let a denote the amount to be stocked. Because the amount to be used X has an exponential distribution, so that π π π π > π = 1 β π π β€ π = 1 β 1 β exp β 4 = exp β 4 So we choose a with P(X>a)=exp(-a/4)=0.05ο a=11.98 (tons). The UNIVERSITY of NORTH CAROLINA at CHAPEL HILL Properties 1. Memoryless: π π > π + π π > π = exp π β π π π>π+π π π>π π+π = 1β 1βexp β π π 1β 1βexp β = π = 1 β πΉ π = π(π > π) 2. Relation with Poisson distribution: Assume a Poisson distribution with Ξ» events per hour, so in t hours, the number of events, Y, follows a Poisson with mean Ξ»t. Now we start at time zero and ask β how long do I have to wait to see the 1st event occur? Let X denote the length of time until 1st event occurs. Ξ»π‘ 0 exp βΞ»π‘ π π > π‘ = π π = 0 ππ π‘βπ πππ‘πππ£ππ 0, π‘ = = exp βΞ»π‘ P(Xβ€ t)=1-exp(- Ξ»π‘) ο π π‘ = 1 exp π 1 β π 0! , π‘ > 0 which means the interval time between two consecutive events in Poisson distribution follows the exponential distribution. The UNIVERSITY of NORTH CAROLINA at CHAPEL HILL Part 3. Gamma: PDF The UNIVERSITY of NORTH CAROLINA at CHAPEL HILL Probability Density Function (PDF) In general, the Gamma density function is given by π π₯ = 1 πΌ β 1exp(β π₯ ), π₯ Ξ πΌ π½πΌ π½ π₯β₯0 0, ππ‘βπππ€ππ π Where the parameters Ξ± and Ξ² are constants (Ξ± >0, Ξ²>0) that determines the shape of the curve. The UNIVERSITY of NORTH CAROLINA at CHAPEL HILL Part 4. Mean and Variance The UNIVERSITY of NORTH CAROLINA at CHAPEL HILL π¬ πΏ = β ππ π π π = ββ π Ξ πΆ π·πΆ β πβπ π βπ πΆ π πππ β π π= π Ξ πΆ π·πΆ π· β πΆ π π πΆ+π π πππ β π π = Ξ πΆ + π π· π π· Ξ πΆ π·πΆ = πΆπ· Similary , we can find π¬(πΏπ) = πΆ(πΆ + π)π·π, so π½(πΏ) = π¬(πΏπ) β π¬π(πΏ) = πΆπ·π. Suppose π = βπΏπ with πΏπ, πΏπ, β¦ , πΏπ being independent Gamma variables with parameters Ξ± and Ξ², then π¬(π) = ππΆπ·, π½(π) = ππΆπ·π. The UNIVERSITY of NORTH CAROLINA at CHAPEL HILL Example 5.11 A certain electronic system has a life length of X1, which has an exponential distribution with a mean of 450 hours. The system is supported by an identical backup system that has a life length of X2. The backup system takes over immediately when the system fails. If the system operate independently, find the probability distribution and expected value for the total life length of the primary and backup systems. Answer: Let Y denote the total life length, Y= X1+X2, where X1 and X2 are Independent exponential random variable with mean Ξ²=450. So Y is a gamma Distribution with Ξ±=2 and Ξ²=450, that is, π π¦ = 1 Ξ 2 450 π¦ , π¦>0 2 π¦ππ₯π β 450 0, ππ‘βπππ€ππ π Then the mean E(Y)=Ξ±Ξ²=2(450)=900. The UNIVERSITY of NORTH CAROLINA at CHAPEL HILL Example 5.12 Suppose that the length of time X needed to conduct a periodic maintenance check on a pathology labβs microscope (known from previous experience) follows a gamma distribution with Ξ±=3 and Ξ²=2 (minutes). Suppose that a new repairperson requires 20 minutes to check a particular microscope. Does this time required to perform a maintenance check seem our of line with prior experience? Answer: so ΞΌ=E(X)=Ξ±Ξ²=6, Ο2=V(X)=Ξ±Ξ²2=12, the standard deviation Ο=3.446, When x=20 minutes required from the repairperson, the deviation is 20-6=14 minutes, Which exceeds the mean 6 by k=14/3.446 standard deviations, so based on the Tschebysheffβs inequality, we have P(|X-6|β₯14)β€(3.446/14)2=0.06, which is really small Probability, so we can say it is out of line with prior experience. The UNIVERSITY of NORTH CAROLINA at CHAPEL HILL Part 3. More Examples The UNIVERSITY of NORTH CAROLINA at CHAPEL HILL Additional Example 1 An insurance policy reimburses dental expense, X, up to a maximum benefit of 250 . The probability density function for X is: where c is a constant. Calculate the median benefit for this policy. Answer: If P(X>a)=1/2, then a is a median. So c=250. As F(x)=1-exp(-x/250), we have 1-exp(-x/250)=1/2 ο x=250[ln(2)] = 173.29 The UNIVERSITY of NORTH CAROLINA at CHAPEL HILL Additional Example 2 Let X be an exponential random variable such that P(X>2) = 2P(X>4). Find the variance of X. Answer: Let the distribution function F π₯ = 1 β exp β π₯ π , based on P(X>2)=2P(X>4), we have 1-F(2)=2(1-F(4))ο 1-(1-exp(-2/ΞΈ))=2(1-(1-exp(-4/ΞΈ))ο exp(-2/ΞΈ)=2exp(-4/ΞΈ)ο -2/ΞΈ=ln(2)-4/ΞΈ ο ΞΈ = 2/ln(2) ο V(X)=[2/ln(2)]2. The UNIVERSITY of NORTH CAROLINA at CHAPEL HILL Additional Example 3 If X has probability density function given by Find the mean and variance. Answer: Change it to the standard form with Ξ±=3, Ξ²=/12, so we can find E(X)=Ξ±Ξ²=3/2, V(X)=Ξ±Ξ²2=3/4. The UNIVERSITY of NORTH CAROLINA at CHAPEL HILL