Transform estimation of parameters for stage-frequency data

• 成果类型: Article
• 署名作者: Hoeting, JA; Tweedie, RL; Olver, CS
• 署名单位: Colorado State University System; Colorado State University Fort Collins; Colorado State University System; Colorado State University Fort Collins; University of Minnesota System; University of Minnesota Twin Cities
• 刊物名称: JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION
• ISSN/ISSBN: 0162-1459
• DOI: 10.1198/016214503000000288
• 发表日期: 2003
• 页码: 503-514
• 关键词: falciparum; SYSTEM; models; times

摘要

We consider multistage development models that occur in the maturation of biological organisms, disease progressions, and industrial processes. The situations that we address are distinguished by the essentially destructive sampling required to assess the stage reached by each individual. We develop robust estimators of stage-dependent maturation rates and overall death rates using method of moments based on Laplace transforms, for which we develop variance estimates under different sampling schemes. We apply these methods to studies of cattle parasite and grasshopper life cycles and show that the Laplace methods compare well with maximum likelihood methods.