Detailed Information

Cited 2 time in webofscience Cited 2 time in scopus
Metadata Downloads

On survival assumptions between integer ages in the theory of competing risks

Authors
Baek, HyeYounKo, BangwonLee, Hangsuck
Issue Date
Dec-2016
Publisher
KOREAN STATISTICAL SOC
Keywords
Absolute decrement rates; Common distribution of decrements; Multiple decrement probabilities; Proportional forces of decrement
Citation
JOURNAL OF THE KOREAN STATISTICAL SOCIETY, v.45, no.4, pp.633 - 646
Journal Title
JOURNAL OF THE KOREAN STATISTICAL SOCIETY
Volume
45
Number
4
Start Page
633
End Page
646
URI
http://scholarworks.bwise.kr/ssu/handle/2018.sw.ssu/5527
DOI
10.1016/j.jkss.2016.05.006
ISSN
1226-3192
Abstract
Calculating survival probabilities based on a multiple decrement life table often requires additional information at arbitrary non-integer ages or conversion formula between multiple decrement probabilities and absolute decrement rates. However, due to the discrete characteristic of the table, it is necessary to impose a certain distributional assumption about the decrements between integer ages such as the uniform distribution of decrements (UDD) or the constant forces (CF) of decrement. In this paper, inspired by the fractional independence (FI) assumption in Willmot (1997), we intend to generalize the commonly used fractional age assumptions and investigate their relationships. In particular, we are interested in relaxing the assumptions while inheriting the same conversion formula from the UDD or the CF assumption. Numerical examples show that our generalization facilitates the investigation of conversion errors caused by incorrectly using the formula when the fractional age assumption is misspecified. (C) 2016 The Korean Statistical Society. Published by Elsevier B.V. All rights reserved.
Files in This Item
There are no files associated with this item.
Appears in
Collections
College of Natural Sciences > ETC > 1. Journal Articles

qrcode

Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.

Related Researcher

Researcher Ko, Bang won photo

Ko, Bang won
College of Natural Sciences (Department of Statistics and Actuarial Science)
Read more

Altmetrics

Total Views & Downloads

BROWSE