13.07.2015 Views

PROCEEDINGS May 15, 16, 17, 18, 2005 - Casualty Actuarial Society

PROCEEDINGS May 15, 16, 17, 18, 2005 - Casualty Actuarial Society

PROCEEDINGS May 15, 16, 17, 18, 2005 - Casualty Actuarial Society

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

476 A NEW METHOD OF ESTIMATING LOSS RESERVESTABLE 6Second Hypothetical Cumulative Incurred Loss Data(in $000s)Development Year (j)AccidentYear (i) 0 1 2 3 4 5 61997 2,237 2,369 2,376 2,376 2,376 2,3761998 2,899 2,942 2,936 2,934 2,9341999 2,225 2,330 2,322 2,3252000 2,145 2,205 2,2072001 1,513 1,4992002 1,<strong>16</strong>8TABLE 7Third Hypothetical Cumulative Incurred Loss Data(in $000s)Development Year (j)AccidentYear (i) 0 1 2 3 4 5 61996 1,076 927 927 951 960 1,087 1,0871997 957 1,193 1,312 1,295 1,220 1,3921998 1,421 1,788 2,086 2,236 2,2521999 1,473 1,910 2,235 2,1922000 1,447 1,976 2,4542001 3,578 3,9112002 4,754Table 7 presents similar challenges as some incremental incurredloss amounts are zero and some are negative. The sequenceof incurred losses generated in 1996 and 1997 appear tohave a pattern distinct from those in subsequent years. In addition,it may be incorrect to assume the accident years all have thesame settlement period, i.e., N depends on i. In such cases wherethere are non-positive incremental incurred losses and/or differentclaim settlement periods, the ultimate incurred loss for accident

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!