23 47 birth processes birth death processes

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: usy t ) 1 = λP0 (t ) − µP1 (t ) ì í î ì í î 0 = −λP0 (t ) + µP1 (t ) ~ Exp(m) ~ Exp(l) Given d π0(t) P0−tλπ+t) + µπ)(t= 1, P0 (t ) + (λ + µ)P0 (t ) = µ. that: = ( ) 0( P1 (t 1 ) dt µ t) π1(t) = P0 (t ) λπ0(t) − µπ1(+ = BY adding both sides of λ + µ the equations d dt Q= P0 (0) − −λ λ µ µ −µ −(λ+µ)t e λ+µ d λ λ ⇒ π0 (t) + π1 (t) = constant = 1 ⇒ (λπ1(t)t= 1 − π0 (t) ) dt (π0 (t) + π1 (t)) = 0 + P1 (0) − P1 (t ) = e− +µ d λ+ λ+µ π0 (t) + (λ + µ)π0 (t) = µµ ⇒ d (e(λ+µ)t π0(t)) = µe(λ+µ)t dt dt µ µ −(λ+µ)t π0(t) Solution = λEquilibrum+distribution + Deviation from the = +µ + (π0(0) − λ µ )e λ π1(t) = λ λ µ + ( (0) − λ+µ )e−(λ+ t equilibrium+withπ1exponentialµ)decay. 23 / 47 Birth Processes Birth-Death Processes Relationship to Markov Chains Linear Birth-Death Processes Examples Example 2. Pure birth process (Poisson process) PoissonProcess. Probabilities λi = λ µ =0 i i = 0, 1 , 2 , . . . πi (0) = 1 0 i=0 i>0 Poisson Process birth probability per time unit is initially the population size • Birth probability per time unit is constant λ constant λ • The population size is initially 0 l 0 l 1 l l 2 l i-1 ... l i All states All states are transient d dt Equations i π (t) = −λπi (t) + λπi−1(t) d dt i>0 > π0(t) Pi (t−= −(tPi (t ) + λPi −1 (t ), i⇒ 0 π0(t) = e−λt = ) λπ0 λ ) P0 (t ) = −λP0 (t ) d λt (e πi(t)) dt = λπi−1(t)eλt ⇒ πi(t) = e−λtλ t 0 πi−1(t )eλ 24 / 47 Birth Processes Birth-Death Processes Relationship to Markov Chains Linear Birth-Death Processes Examples Poisson Process. Probabilities Equations Pi (t ) = −λPi (t ) + λPi −1 (t ), i > 0 P0 (t ) = −λP0 (t ) ⇒ P0 (t ) = e−λt d λt [e Pi (t )] = λPi −1 (t )eλt ⇒ Pi (t ) = e−λt λ dt t P1 (t ) = e−λt λ t Pi −1 (t )eλt dt 0 e−λt eλt dt = e−λt (λt ) 0 (λt )i −λt e i! Number of births in interval (0, t ) ∼ Poisson(λt ). Recursively: Pi (t ) = 25 / 47 Example Processes Relationship toprocess Birth-Death 1. Pure death Markov Chains Birth Processes Linear Birth-Death Processes Examples λi = 0 Pure Death Process. i = 0, 1, 2, . . . Probabilities µ = iµ i πi(0) = all individuals have the same the Pure Death Process mortality rate µ • All the individuals have the same mortality rate µ 1 0 i= i= system starts • The population size is initially n 0 1 m 2 2m n-1 ... 3m (n-1) m Sta oth n nm State 0 is an absorbing state. The rest are transient. Equations d dt πn(t) = −nµπn(t) ⇒ P(t ) = −nµPn (t ) n d Pi (tdt =i(it) 1)µPii +t1)µπiPi (tt) − = 0, i (t) n − 1 ) π ( + = ( +1 ( ) − i µ +1( ), i iµπ . . . , d iµt dt (e πi (t)) = (i + 1)µπi+1 (t)eiµt ⇒ πn(t) i = 0, 1 , . . πi (t) 26 / 47(i = Birth Processes Birth-Death Processes Relationship to Markov Chains Linear Birth-Death Processes Examples Pure Death Process. Probabilities Equations Pn (t ) = −nµPn...
View Full Document

This document was uploaded on 01/20/2014.

Ask a homework question - tutors are online