{[ promptMessage ]}

Bookmark it

{[ promptMessage ]}

FinalexamACTL2003-Yr2009-Sol

FinalexamACTL2003-Yr2009-Sol - ACTL 2003/5103 Final...

This preview shows pages 1–4. Sign up to view the full content.

ACTL 2003/5103 Final Examination-Sample Solutions-year 2009 Question 1 (a) . Sol. { 1 , 5 } transient { 3 } recurrent { 2 , 4 } recurrent (b) Sol. P T = ± 1 5 4 5 1 2 0 S = ( I - P T ) ( - 1) = ± 4 5 - 4 5 - 1 2 1 - 1 = 1 4 5 1 2 4 5 4 5 - 4 5 - 1 2 1 = ± 5 2 2 5 4 2 (c) Sol. Pr ( X 2 = 3) = 5 X i =1 Pr ( X 2 = 3 | X 0 = i ) Pr ( X 0 = i ) = 5 X i =1 P (2) i 3 i 15 = 1 15 ( 1 5 0 0 0 4 5 )(0 0 1 0 1 2 ) T + 2 15 (0 1 3 0 2 3 0)(0 0 1 0 1 2 ) T + 3 15 (0 0 1 0 0)(0 0 1 0 1 2 ) T + 4 15 (0 1 2 0 1 2 0)(0 0 1 0 1 2 ) T + 5 15 ( 1 2 0 1 2 0 0)(0 0 1 0 1 2 ) T = 1 15 × 2 5 + 0 + 3 15 + 0 + 5 15 × 1 2 = 59 150 = 0 . 39 An easier way is to use the fact that 3 is absorbing ( so p (2) 33 = 1), the tran- sition probability from a recurrent state to a transient state is always 0, (so 1

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
p (2) 23 = 0 = p (2) 43 ). (d) Sol. The chain is not ergodic due to any of the following reasons: it is reducible and it is not positive recurrent. (e) Sol. Since when the chain enters in class { 2 , 4 } it will always stay in a state within the class. Go starting from state 2, the Markov chain is same as a Markov chain starting from state 2 with state space { 2 , 4 } and transition probability matrix ± 1 3 2 3 1 2 1 2 . Note the new Markov chain is ergodic and therefore the limiting proba- bilities lim n →∞ p 24 and lim n →∞ p 42 exist. Let x = lim n →∞ p 42 and y = lim n →∞ p 24 . Then x = 1 3 x + 1 2 y y = 2 3 x + 1 2 y x + y = 1 Solving the equations gives x = 3 7 and y = 4 7 . So lim n →∞ p 24 = 4 7 . Question 2 (a) (i) Sol. Let X n denote the state of a policyholder in year n . { X n } is a Markov chain with state space { 1 , 2 , 3 } . 0 . 9 0 . 1 0 0 . 7 0 . 2 0 . 1 0 0 . 7 0 . 3 2
(ii) Sol. Solving π 1 = 0 . 9 π 1 + 0 . 7 π 2 π 2 = 0 . 1 π 1 + 0 . 2 π 2 + 0 . 7 π 3 π 3 = 0 . 1 π 2 + 0 . 3 π 3 π 1 + π 2 + π 3 = 1 gives π 1 = 49 57 , π 2 = 7 57 , and π 3 = 1 57 . The long-run average premium a policyholder pays per year=100 π 1 + 200 π 2 + 300 π 3 = 115 . 79 (b) Sol. Method 1: Let the state equal to the capital at the end of a year. The state space S = { 0 , 1 , 2 } . The one-step transition probability matrix is 1 0 0 1 3 1 3 1 3 1 4 1 4 1 2 Note the statement “after receiving the capital injection at the beginning of a year, the capital is 3 million” is equivalent to“ at the end of last year, the capital is 2 million”. Since 0 is absorbing, so whenever the chain reaches state 0, it will always stay there. So the desired probability is

This preview has intentionally blurred sections. Sign up to view the full version.

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

{[ snackBarMessage ]}

Page1 / 13

FinalexamACTL2003-Yr2009-Sol - ACTL 2003/5103 Final...

This preview shows document pages 1 - 4. Sign up to view the full document.

View Full Document
Ask a homework question - tutors are online