Take the 2-minute tour ×
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

What type of Markov process relates to an inhomogeneous Poisson process?

A homogeneous Poisson process-- one where the rate, $\lambda$, is constant-- is a pure birth continuous time Markov chain (with a constant birth rate). That is, if $q_{ij}$ is the transition rate then $q_{ij} = 0 $ unless $j=i+1$ i which case $q_i,i+1 = \lambda$

What happens when $\lambda = \lambda \left(t\right)$? Does, $q \rightarrow q\left(t\right)$? And, how is this represented in a transition matrix?

For a homogeneous Poisson process, the matrix of transition probabilities for the corresponding Markov process is one that is zero every where except the band immediately above the diagonal, all entries of which are $\lambda$.

share|improve this question
add comment

1 Answer

up vote 1 down vote accepted

Well, let's try this... Let $\{N_{t}\}_{t\geq0}$ be an inhomogeneous Poisson process with rate function $\{\lambda(t)\}_{t\geq0}$. Let's try to compute the transition probabilites.

By definition of an inhomogeneous Poisson process, we have

$$P\{N(t+h)-N(t)=1\}=\lambda(t)h+o(h)$$ $$P\{N(t+h)-N(t)>1\}=o(h)$$

Since $\{N_{t}\}_{t\geq0}$ has independent increments by definition, we have

$$P\{N(t+h)-N(t)=1|N(t)=k\}=\lambda(t)h+o(h)$$ $$P\{N(t+h)-N(t)>1|N(t)=k\}=o(h)$$

for any $t\geq0$, any $k\in\mathbb{N}_{0}$ and "small" $h$.

Dividing previous equations by $h$, and letting $h\rightarrow 0$, we get

$$\lim_{h\rightarrow0}\frac{1}{h}P\{N(t+h)-N(t)=i+1|N(t)=i\}=\lambda(t)$$ $$\lim_{h\rightarrow0}\frac{1}{h}P\{N(t+h)-N(t)=j|N(t)=i\}=0$$ for $j>i+1$ or $j<i$.

share|improve this answer
    
I see, so it is the case that the probability of transition still tracks $\lambda$. Thanks. –  mac389 Jun 15 '13 at 14:29
    
What about the case where $\lambda\left(t\right)$ is periodic and the period $\geq 10h$? –  mac389 Jun 15 '13 at 14:30
    
Well, the reasoning was sort of independent of the form of $\lambda(t)$, so the answer would be the same, I guess... –  Stojan Jovanović Jun 15 '13 at 20:39
add comment

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.