# Hierarchical model

• October 25th 2008, 02:48 AM
approx
Hierarchical model
I'm having some trouble with hierarchical models.

Suppose that a company has determined that the number of jobs per week, N, varies from week to week and has a Poisson distribution with mean $\lambda$. The number of hours to complete each job, $Y_{i}$, is gamma distributed with parameters $\alpha$ and $\beta$. The total time to complete all jobs in a week is: $T = \sum_{i=1}^{N}{Y_i}$

Note that T is the sum of a random number of random variables.

a)
What is $E(T\mid N=n)$?

The correct answer is $n\alpha\beta$, but I don't know how to get there? Is it correct to do like this?

$E(T\mid N=n) = E( \sum_{i=1}^{N}{Y_i}\mid N=n)$ =
$\sum_{i=1}^{N}E(Y_i\mid N=n)$ =
$\sum_{i=1}^{n} E(Y_i)$

and because $Y_i$ is gamma distributed =>
$
E(Y_i) = \alpha \beta$

$\sum_{i=1}^{n}{\alpha \beta } = n\alpha \beta$

b) $E(T) = E(\sum_{i=1}^{N}{Y_i})$ =

$\sum_{i=1}^{N} E(Y_i)$

$= \sum_{i=1}^{N} \alpha \beta$

The correct answer is $= \lambda \alpha \beta$, but I don't know how to get there. I don't really know how to think when it comes to hierarchical models. It's kind of difficult, I think. Thanks in advance for your help.