APC Accounting/Multiple Classification Model, Part I: Model Identification and Estimation Using the Intrinsic Estimator
Year: 2013
Type: book-chapter
Abstract: This chapter focuses on an innovative approach to the model identification and estimation problem within the linear/generalized linear models framework for age-period-cohort (APC) analysis. Incorporating recent methodological developments on estimable functions in the context of the APC accounting model, a new method of estimation termed the Intrinsic Estimator (IE) is proposed and compared to the conventional coefficient constraints approach, that is, the constrained generalized linear model. To date, the IE has been empirically applied in a number of studies in epidemiology, demography, and social science and has yielded sensible results on age, period, and cohort trends in a variety of topics, including both general and cause-specific mortality rates. The chapter defines the IE, describes its statistical properties, and illustrates the robustness of these properties via a comparison with age, period, and cohort estimates from an empirical application of a hierarchical APC model and from a numerical simulation analysis.
Source: Chapman & Hall/CRC eBooks
DOI: 10.1201/b13902-5
Authors: Yang Yang, Kenneth C. Land
Language: English
Subfield: Statistics and Probability
Field: Mathematics
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W3004924126