Computation of age-specific hiv incidence estimates using the em algorithm

Ian C. Marschner*

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

2 Citations (Scopus)

Abstract

The fitting of age-dependent HIV incidence models to AIDS data is a computationally intensive task, particularly when allowance is made for non-proportional dependence of the infection rate on age. This paper presents a computational alternative to a very intensive method described by Rosenberg (1994). Our approach is to use the EM algorithm on a discretized form of the model used by Rosenberg (1994). The EM approach has certain attractive features including ease of implementation and flexibility of model specification. It also conveniently generalizes to allow smoothed estimation and less detailed forms of age-specific AIDS data.

Original languageEnglish
Pages (from-to)299-312
Number of pages14
JournalJournal of Statistical Computation and Simulation
Volume53
Issue number3-4
DOIs
Publication statusPublished - 1995
Externally publishedYes

Keywords

  • Age-dependent incidence
  • AIDS/HIV
  • Back-projection
  • EM algorithm
  • Log-linear model
  • Multiplicative model
  • Smoothing

Fingerprint

Dive into the research topics of 'Computation of age-specific hiv incidence estimates using the em algorithm'. Together they form a unique fingerprint.

Cite this