STAT3801_2011Unit2[1]

# STAT3801_2011Unit2[1] - THE UNIVERSITY OF HONG KONG...

This preview shows pages 1–6. Sign up to view the full content.

THE UNIVERSITY OF HONG KONG DEPARTMENT OF STATISTICS AND ACTUARIAL SCIENCE STAT3801 ADVANCED LIFE CONTINGENCIES Unit 2 2010 - 2011 2 nd Semester Multiple life functions There must be a ﬁnite set of members for each member deﬁne future lifetime random variable Status survival can be determined at any future time 2.1 Joint distribution of future lifetimes Example 9.2.1 f T ( x ) T ( y ) ( s, t ) = ( 0 . 0006( t - s ) 2 0 < s < 10 , 0 < t < 10 0 elsewhere Required to determine Joint d. f. of T ( x ) and T ( y ) Marginal p. d. f., d. f., s p x and μ ( x + s ) for T ( x ) Correlation coeﬃcient of T ( x ) and T ( y )

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

View Full Document
2 Sample space of T ( x ) and T ( y ) 6 - 0 T ( x ) T ( y ) I II III IV 10 10
3 Region I F T ( x ) T ( y ) ( s, t ) = Pr [ T ( x ) s and T ( y ) t ] = Z s -∞ Z t -∞ f T ( x ) T ( y ) ( u, v ) dv du = Z s 0 Z t 0 0 . 0006( v - u ) 2 dv du = Z s 0 0 . 0002[( t - u ) 3 + u 3 ] du = 0 . 00005[ s 4 + t 4 - ( t - s ) 4 ] 0 < s 10 , 0 < t 10 Regions II, III and IV F T ( x ) T ( y ) ( s, t ) = F T ( x ) T ( y ) ( s, 10) = F T ( x ) ( s ) = 1 2 + 0 . 00005[ s 4 - (10 - s ) 4 ] in Region II = F T ( x ) T ( y ) (10 , t ) = F T ( y ) ( t ) = 1 2 + 0 . 00005[ t 4 - (10 - t ) 4 ] in Region IV = 1 in Region III Marginal distribution and density of T ( x ) F T ( x ) T ( y ) ( s, 10) = F T ( x ) ( s ) = 0 s 0 1 2 + 0 . 00005[ s 4 - (10 - s ) 4 ] 0 < s 10 1 s > 10 f T ( x ) ( s ) = F 0 T ( x ) ( s ) = ( 0 . 0002[ s 3 + (10 - s ) 3 ] 0 < s < 10 0 elsewhere

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

View Full Document
4 Survival probability and force of mortality s p x = 1 - F T ( x ) ( s ) = 1 2 + 0 . 00005[(10 - s ) 4 - s 4 ] 0 < s 10 = 0 s > 10 μ ( x + t ) = f T ( x ) ( t ) 1 - F T ( x ) ( t ) = 0 . 0002[ s 3 + (10 - s ) 3 ] 1 2 + 0 . 00005[(10 - s ) 4 - s 4 ] 0 < s 10 Correlation coeﬃcient of T ( x ) and T ( y ) E [ T ( x )] = Z 10 0 s (0 . 0002)[ s 3 + (10 - s ) 3 ] ds = 5 = E [ T ( y )] E [ T ( x ) 2 ] = Z 10 0 s 2 (0 . 0002)[ s 3 + (10 - s ) 3 ] ds = 110 3 = E [ T ( y ) 2 ] V ar [ T ( x )] = 35 3 = V ar [ T ( y )] E [ T ( x ) T ( y )] = Z 10 0 Z 10 0 s t (0 . 0006)( t - s ) 2 ds dt = 50 3 Cov [ T ( x ) , T ( y )] = E [ T ( x ) T ( y )] - E [ T ( x )] E [ T ( y )] = - 25 3 ρ T ( x ) T ( y ) = Cov [ T ( x ) , T ( y )] σ T ( x ) σ T ( y ) = - 25 3 35 3 = - 5 7
5 2.2 Joint survival function s T ( x ) T ( y ) ( s, t ) = Pr [ T ( x ) > s and T ( y ) > t ] Sample space of future lifetime random variables T ( x ) and T ( y ) 6 - 0 T ( x ) T ( y ) Region A ( s, t ) @ @ @ R Region B s t Joint survival function For 0 < s < 10 and 0 < t < 10 , s T ( x ) T ( y ) ( s, t ) = Pr [ T ( x ) > s T ( y ) > t ] = Z s Z t f T ( x ) T ( y ) ( u, v ) dv du = Z 10 s Z 10 t 0 . 0006( v - u ) 2 dv du = Z 10 s 0 . 0002[(10 - u ) 3 - ( t - u ) 3 ] du = 0 . 00005[(10

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.

## This note was uploaded on 07/30/2011 for the course STAT 3801 taught by Professor Kc during the Fall '11 term at HKU.

### Page1 / 24

STAT3801_2011Unit2[1] - THE UNIVERSITY OF HONG KONG...

This preview shows document pages 1 - 6. Sign up to view the full document.

View Full Document
Ask a homework question - tutors are online