3 views

Uploaded by Amit Patil

no

save

- IJCMS-V3I3-2012-01
- MST121 Fall Exam 2007-2008
- 465_SampleQuiz_2
- FORECAST OF SPORTS RESULTS
- sta255_Lecture3
- Hw2 Solution
- MIL-STD-105E_text.pdf
- Operations Research 2 Mid Online Bits
- S2 Jan 2008
- Mathematics Report
- Week3.pdf
- Probabilit1
- Additional Mathematics Project 2016
- HW5 Selected Solutions
- s1 Past Paper Questions
- A Comparative Analysis of QADA-KF with JPDAF for Multitarget Tracking in Clutter
- upfal
- Solomon E
- AnbarasanThangaveluMFKM2007
- Aim Report Final43960
- 2014 Sd Mechanical Contact Best Practices
- Create Manikin
- Electrothermal Simulation
- Intro and Course Flow of delmia

You are on page 1of 14

**Manual for SOA Exam MLC.
**

Chapter 11. Poisson processes.

Section 11.5. Nonhomogenous Poisson processes

c

2009.

Miguel A. Arcones. All rights reserved.

Extract from:

”Arcones’ Manual for SOA Exam MLC. Fall 2009 Edition”,

available at http://www.actexmadriver.com/

1/14

c

2009.

Miguel A. Arcones. All rights reserved.

Manual for SOA Exam MLC.

N(t1 ). (iii) For each 0 ≤ t1 < t2 < · · · < tm . .v. . Nonhomogenous Poisson processes Definition 1 The counting process {N(t) : t ≥ 0} is said to be a nonhomogenous Poisson process with intensity function λ(t). (ii) For each R t t > 0. Arcones. . t ≥ 0. . N(t2 ) − N(t1 ). N(tm ) − N(tm−1 ) are independent r. Miguel A. All rights reserved.5. if (i) N(0) = 0. Section 11. 2/14 c 2009. . Nonhomogenous Poisson processes. Poisson processes. N(t) has a Poisson distribution with mean m(t) = 0 λ(s) ds. Manual for SOA Exam MLC.Chapter 11.’s.

.v.5. 3/14 c 2009. if (i) N(0) = 0. t ≥ 0. . N(t2 ) − N(t1 ). Nonhomogenous Poisson processes Definition 1 The counting process {N(t) : t ≥ 0} is said to be a nonhomogenous Poisson process with intensity function λ(t). . Poisson processes. I An nonhomogeneous Poisson process with λ(t) = λ. Manual for SOA Exam MLC. N(t1 ). is a regular Poisson process. (ii) For each R t t > 0. Nonhomogenous Poisson processes. N(t) has a Poisson distribution with mean m(t) = 0 λ(s) ds.Chapter 11. (iii) For each 0 ≤ t1 < t2 < · · · < tm . Miguel A. .’s. . Arcones. Section 11. N(tm ) − N(tm−1 ) are independent r. All rights reserved. for each t ≥ 0.

Miguel A. . for each t ≥ 0.Chapter 11. Poisson processes. if (i) N(0) = 0. . N(tm ) − N(tm−1 ) are independent r. N(t2 ) − N(t1 ). . Arcones.5. Nonhomogenous Poisson processes. I m(t) is the mean value function of the non homogeneous Poisson process. (iii) For each 0 ≤ t1 < t2 < · · · < tm .’s. Nonhomogenous Poisson processes Definition 1 The counting process {N(t) : t ≥ 0} is said to be a nonhomogenous Poisson process with intensity function λ(t). . N(t) has a Poisson distribution with mean m(t) = 0 λ(s) ds. All rights reserved. I An nonhomogeneous Poisson process with λ(t) = λ. Manual for SOA Exam MLC. . 4/14 c 2009. (ii) For each R t t > 0. is a regular Poisson process. t ≥ 0.v. N(t1 ). Section 11.

5/14 c 2009. . Nonhomogenous Poisson processes. All rights reserved. Manual for SOA Exam MLC.Chapter 11. I Section 11. but not necessarily stationary.5. Arcones. Poisson processes. Miguel A. The increments of an nonhomogeneous Poisson process are independent.

Poisson processes. Manual for SOA Exam MLC.5. Miguel A. I A nonhomogeneous Poisson process is a Markov process. Nonhomogenous Poisson processes. . All rights reserved. I The increments of an nonhomogeneous Poisson process are independent. Arcones. 6/14 c 2009. but not necessarily stationary. Section 11.Chapter 11.

Arcones. Miguel A. . Manual for SOA Exam MLC.Chapter 11. I A nonhomogeneous Poisson process is a Markov process. I For each 0 ≤ s < t.5. but not necessarily stationary. 7/14 c 2009. All rights reserved. I The increments of an nonhomogeneous Poisson process are independent. N(t) R t − N(s) has Poisson distribution with mean m(t) − m(s) = s λ(x) dx. Nonhomogenous Poisson processes. Section 11. Poisson processes.

(km −km−1 )! ··· 8/14 c 2009. I A nonhomogeneous Poisson process is a Markov process. Miguel A.Chapter 11. . I The increments of an nonhomogeneous Poisson process are independent. . N(tm ) − N(tm−1 ) = km − km−1 } = e −m(t1 ) (m(t1 ))k1 e −(m(t2 )−m(t1 )) (m(t2 )−(m(t1 ))k2 −k1 k1 ! (k2 −k1 )! e −(m(tm )−m(tm−1 )) (m(tm )−m(tm−1 ))km −km−1 . . I For each 0 ≤ t1 < t2 < · · · < tm and each integers k1 . N(tm ) = km } = P{N(t1 ) = k1 .5. . N(t2 ) = k2 . All rights reserved. Manual for SOA Exam MLC. . . . P{N(t1 ) = k1 . . N(t) R t − N(s) has Poisson distribution with mean m(t) − m(s) = s λ(x) dx. . Arcones. km ≥ 0. Poisson processes. . . I For each 0 ≤ s < t. . . Nonhomogenous Poisson processes. Section 11. but not necessarily stationary. N(t2 ) − N(t1 ) = k1 − k1 .

25. 3].Chapter 11. (3. . . . Arcones. Section 11. Nonhomogenous Poisson processes. . Miguel A. All rights reserved. 4]. . 9/14 c 2009. 2]. Find the probability that the number number of observed occurrences in the time period (1. λ(t) = 3 if t is in (0. 3] is more than two.5. (2. Manual for SOA Exam MLC. . Example 1 For a nonhomogenous Poisson process the intensity function is given by ( 5 if t is in (1. Poisson processes. . 1].

25) has a Poisson distribution with mean Z 3 Z 2 Z 3 m(3) − m(1. 3]. All rights reserved. Section 11. 2]. . Nonhomogenous Poisson processes.25) = λ(t) dt = 5 dt + 3 dt = 6.Chapter 11. λ(t) = 3 if t is in (0. . Poisson processes.25 2 Hence. (2.25 1. Find the probability that the number number of observed occurrences in the time period (1.9642515816.5. Manual for SOA Exam MLC. 10/14 c 2009. P{N(3) − N(1. . Example 1 For a nonhomogenous Poisson process the intensity function is given by ( 5 if t is in (1.25. .75 + (6. . Solution: N(3) − N(1.75 (1 + 6. Miguel A. . 4].75)2 /2) =0. (3. . 3] is more than two.25) > 2} = 1 − e −6. 1]. Arcones.75. 1.

(n − 1)! (n − 1)! If lim m(t) < ∞. 1) ≥ lim m(t)} = 0 t→∞ t→∞ and Sn is a r. e −m(t) (m(t))n−1 λ(t) .v. Section 11. Poisson processes. Let Sn be the time of the n–th occurrence. 11/14 . Nonhomogenous Poisson processes. Miguel A. 1) ≥ lim m(t)} > 0 t→∞ and density of its continuous part fSn (t) = c 2009.Chapter 11. 1) ≥ m(t)}. then Sn is a mixed r. ∞]. If lim m(t) = ∞.v. t ≥ 0. P e −m(t) (m(t))j P{Sn > t} = P{N(t) ≤ n − 1} = n−1 j=0 j! = P{Gamma(n. Then. then t→∞ P{Sn = ∞} = lim P{Sn > t} = P{Gamma(n. t ≥ 0. (n−1)! Manual for SOA Exam MLC. The density of Sn is fSn (t) = e −m(t) (m(t))n−1 m0 (t) (m(t))n−1 λ(t) = e −m(t) .v. with t→∞ P{Sn = ∞} = P{Gamma(n. with values in [0. Arcones.5. All rights reserved. Sn is an extended r.

For 0 ≤ s ≤ t. Let Tn = Sn − Sn−1 be the n–th interarrival time. Section 11. Nonhomogenous Poisson processes. . For a non–homogeneous Poisson process {Tn }∞ n=1 are not necessarily independent r.’s. P{Tn+1 > t|Sn = s} = P{Sn+1 > s + t|Sn = t} =P{N(s + t) = n + 1|Sn = t} = P{N(s + t) − N(s) = 1|Sn = t} =P{N(s + t) − N(s) = 1} = e −(m(s+t)−m(s)) . Notice Sn = t depends on the non–homogeneous Poisson process until time t. Poisson processes. Hence. {N(s + t) − N(s) = 1} and {Sn = t} are independent. Arcones. Miguel A. 12/14 c 2009. Manual for SOA Exam MLC. All rights reserved.5.Chapter 11.v.

Arcones. Manual for SOA Exam MLC. If S5 = 2. All rights reserved.5.Chapter 11. the intensity function is given by ( t for 0 < t ≤ 4. Poisson processes. . λ(t) = 4 for 10 < t. Example 2 For a non–homogenous Poisson process. calculate the probability that S6 > 5. 13/14 c 2009. Section 11. Nonhomogenous Poisson processes. Miguel A.

14/14 .Chapter 11. c 2009.5. Manual for SOA Exam MLC. calculate the probability that S6 > 5. Section 11. Arcones. Miguel A. Nonhomogenous Poisson processes. Poisson processes. λ(t) = 4 for 10 < t. 4 Hence. Solution: We have that P{S6 > 5|S5 = 2} = P{N(5) = 5|S5 = 2} = P{N(5) − N(2) = 0|S5 = 2} = P{N(5) − N(2) = 0} and Z m(5) − m(2) = 5 Z Z t dt + λ(t) dt = 2 4 2 5 4 dt = 10. All rights reserved. If S5 = 2. the intensity function is given by ( t for 0 < t ≤ 4. P{S6 > 5|S5 = 2} = P{N(5) − N(2) = 0} = e −10 . Example 2 For a non–homogenous Poisson process.

- IJCMS-V3I3-2012-01Uploaded byKurumeti Naga Surya Lakshmana Kumar
- MST121 Fall Exam 2007-2008Uploaded byyAlhashim
- 465_SampleQuiz_2Uploaded byAnh Thu Tran Xuan
- FORECAST OF SPORTS RESULTSUploaded bycika-dule
- sta255_Lecture3Uploaded byIEyra ShaHera
- Hw2 SolutionUploaded byaswaniexpress
- MIL-STD-105E_text.pdfUploaded byAntonios Papadopoulos
- Operations Research 2 Mid Online BitsUploaded bytslnarayana
- S2 Jan 2008Uploaded byG-mi Jimii Kaku
- Mathematics ReportUploaded byAmrit Sonia
- Week3.pdfUploaded byKai Liu
- Probabilit1Uploaded byfarhanshakeel
- Additional Mathematics Project 2016Uploaded by邓心悦
- HW5 Selected SolutionsUploaded byAyan Sinha
- s1 Past Paper QuestionsUploaded byMohammed
- A Comparative Analysis of QADA-KF with JPDAF for Multitarget Tracking in ClutterUploaded byMia Amalia
- upfalUploaded bypreethi
- Solomon EUploaded bysorac