Scribd
Upload
60 views4 pages

406 Mark Scheme

1. This document presents five parts of the C.A.T MARKING SCHEME. It discusses techniques for calculating probabilities, moment generating functions, and time to extinction in pure death processes and birth-death processes. Key formulas are derived for probability generating functions, mean and variance of waiting times, solutions to differential equations, and probability distributions of time to extinction.

Uploaded by

Patrick Mugo
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
60 views4 pages

406 Mark Scheme

1. This document presents five parts of the C.A.T MARKING SCHEME. It discusses techniques for calculating probabilities, moment generating functions, and time to extinction in pure death processes and birth-death processes. Key formulas are derived for probability generating functions, mean and variance of waiting times, solutions to differential equations, and probability distributions of time to extinction.

Uploaded by

Patrick Mugo
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 4

C.A.

T MARKING SCHEME
(1) Pr(X = n) = Pr(X > n 1) Pr(X > n)
Hence multiplying through by S n and summing over all possible values of n we get
X X X
n n
Pr(X = n)S = Pr(X > n 1)S Pr(X > n)S n
n=1 n=1 n=1

Now for each term in the expression we have


X
Pr(X = n)S n = G(s) Pr(X = 0)
n=1
X

Pr(X > n 1)S n = Pr(X > 0)S 1 + Pr(X > 1)S 2 + ... + Pr(X > n 1)S n + ...
n=1
= S(Pr(X > 0) + Pr(X > 1)S 1 + ... + Pr(X > n 1)S n1 + ...
= S(s)
X

Pr(X > n)S n = Pr(X > 1)S 1 + Pr(X > 2)S 2 + ... + Pr(X > n)S n + ...
n=1
= (s) Pr(X > 0)
Therefore
G(s) Pr(X = 0) = S(s) ((s) Pr(X > 0))
G(s) Pr(X = 0) Pr(X > 0) = S(s) (s)

Now Pr(X = 0) + Pr(X > 0) = 1

G(s) 1 = (s)(s 1)

G(s) 1
(s) =
s1
when s=1, G(1) =1 hence (1) = 0

(2) (i) H(s) = p0 + p1 s , Pr(Zn = 0) = Hn (0) where Hn (s) is the p.g.f of Zn


H(0) = p0
H2 (0) = H(H(0)) = p0 + p1 p0
H3 (0) = H(H2 (0)) = p0 + p1 (p0 + p1 p0 )
= p0 + p1 p0 + p21 p0
H4 (0) = H(H3 (0)) = p0 + p1 (p0 + p1 p0 + p21 p0 )
= p0 + p1 p0 + p21 p0 + p31 p0

1
In general then

Hn (0) = H(Hn1 (0)) = p0 + p1 (p0 + p1 p0 + ... + pn2


1 p0 )
= p0 + p1 p0 + p21 p0 + ... + pn1
1 p0
n
1 p1
= p0
1 p1
= 1 pn1 n = 1, 2, ...

(ii)

H(s) = s
p0 + p1 s = s
p0 = s(1 p1 )
1 = s

Hence the probability of extinction with one ancestor is 1. For the two ancestors acting
independently, the probability is 1 1 = 1

(3) The time between successive births is a random variable having exponential distribution
having rate paramter equal to current birth rate.
Let Tj be the time between j th and j + 1 th births; j=1,2,..,N-1. The waiting time until
the (N-1)-th birth is the sum
NX2
W = Tj
j=1

W is the sum of independent exponential random variables;

X
N 2
E[W ] = E[Tj ]
j=1
N 1
=
k(N K)

X
N 2
Var[W ] = Var[Tj ]
j=1
 2
N 1
=
k(N K)

(4) Using the method of momemts


X X X X
np0n (t) = n ( + n)pn (t) + npn1 (t) + n(n + 1)pn+1 (t)
n=1 n=1 n=1 n=1

2
Now
X
np0n (t) = M10 (t)
n=1
X X X
n ( + n)pn (t) = npn (t) n2 pn (t)
n=1 n=1 n=1
= M10 (t) M20 (t)
X X
npn1 (t) = [(n 1) + 1]pn1 (t) = M10 (t) +
n=1 n=1
X X
n(n + 1)pn+1 (t) = [(n + 1) 1](n + 1)pn+1 (t)
n=1 n=1
X X
= (n + 1)2 pn+1 (t) (n + 1)pn+1 (t)
n=1 n=1
= (M20 (t) p1 (t)) 0
(M1 (t) p1 (t))

Hence

M10 (t) = M10 (t) M20 (t) + M10 (t) + + (M20 (t) p1 (t)) (M10 (t) p1 (t))
= M10 (t) M10 (t) +

Multiplying the final expression by integrating factor e(+)t;

M10 (t)e+)t + e(+)t( + )M10 (t) = e(+)t


dM10 (t)e(+)t
= e+)t
dt
Z Z
dM10 (t)e(+)t
dt = e(+)tdt
dt
(+)t
M10 (t)e(+)t = e +c
+
(+)t
M10 (t) = e(+)t e + e(+)tc
+

= + e(+)t c
+
at time t= 0, M1 (0) = i

M10 (0) = + e(+)0 c
+

i = +c
+
( + )i
c =
+

3
Hence
( + )i
M10 (t) = + e(+)t
+ +

(5) The time to extinction ofr pure death process is equivalent to having no one in the system.
For T we have
Pr(T t) = Pr(X(t) = 0)
Now
N
Pr(X(t) = 0) = p0 (t) = 1 et
Hence
N
Pr(T t) = 1 et
N
F (t) = 1 et

Differentiating this function with respect to t

dF (t) N 1
= Net 1 et
dt
N 1
f (t) = Net 1 et

T is the sum of N independent random variables having exponential distribution with rate
parameter equal to the current death rate.

You might also like