Location
Rideau Salon, Ottawa Marriott Hotel
Event Website
http://sociology.uwo.ca/cluster/en/projects/knowledge_mobilization/2015/2015_conference/index.html#2015 Conference
Start Date
18-3-2015 8:30 AM
End Date
18-3-2015 5:00 PM
Description
Multistate Analysis of Life Histories with R
Frans Willekens, Netherlands Interdisciplinary Demographic Institute and Max Planck Institute for Demographic Research
Workshop Venue: Rideau Salon
Workshop Outline
A. Introduction
Multistate models describe the life course in terms of transitions individuals experience as they go through stages of life and move between states. These states may represent health states, family status, occupation, place of residence, education or other domains of life. Multistate models have been successfully used in a wide variety of applied sciences. The most fruitful areas of application are health sciences, demography and economics. Important examples of applications of multistate models are stem cell transplantation (with disease relapse and death as endpoints and graft-versus-host disease as intermediate states), estimation of healthy life expectancy, marital careers, migration histories, and participation in the labor market. These applications have in common a fundamental interest in competing risks, event histories and state sequences. The subject of the workshop is the modeling of life histories.
Multistate analysis of life histories with R is an introduction to multistate event history analysis. It is an extension of survival analysis, in which a single terminal event (endpoint) is considered and the time-to-event is studied. Life histories are modeled as realizations of continuous-time Markov processes (and extensions). The statistical theory of counting processes emerged as the dominant theory for estimating transition rates from data on event counts and populations at risk. Non-parametric and parametric methods have been developed.
In recent years, software packages for multistate modeling have become available. R is the language of choice and the Comprehensive R Archive Network (CRAN) is the main repository. The packages are free and the source code is available in CRAN. The packages include survival, eha, Epi, mvna, etm, mstate, msm, Biograph, MicSim andTraMineR. For a recent review, see Willekens and Putter (2014) Software for multistate analysis, Demographic Research 31(14):381-420, and the CRAN Task View on Survival Analysis In 2011, the Journal of Statistical Software published a special issue on multistate modeling (H. Putter, editor). Multistate modeling is an active area of research across disciplines. The research benefits from the current interest in prognostic modeling (of outcomes of health conditions and behavior/lifestyle) and predictive analytics.
Included in
Preconference Training Workshop: Multistate analysis of life histories with R
Rideau Salon, Ottawa Marriott Hotel
Multistate Analysis of Life Histories with R
Frans Willekens, Netherlands Interdisciplinary Demographic Institute and Max Planck Institute for Demographic Research
Workshop Venue: Rideau Salon
Workshop Outline
A. Introduction
Multistate models describe the life course in terms of transitions individuals experience as they go through stages of life and move between states. These states may represent health states, family status, occupation, place of residence, education or other domains of life. Multistate models have been successfully used in a wide variety of applied sciences. The most fruitful areas of application are health sciences, demography and economics. Important examples of applications of multistate models are stem cell transplantation (with disease relapse and death as endpoints and graft-versus-host disease as intermediate states), estimation of healthy life expectancy, marital careers, migration histories, and participation in the labor market. These applications have in common a fundamental interest in competing risks, event histories and state sequences. The subject of the workshop is the modeling of life histories.
Multistate analysis of life histories with R is an introduction to multistate event history analysis. It is an extension of survival analysis, in which a single terminal event (endpoint) is considered and the time-to-event is studied. Life histories are modeled as realizations of continuous-time Markov processes (and extensions). The statistical theory of counting processes emerged as the dominant theory for estimating transition rates from data on event counts and populations at risk. Non-parametric and parametric methods have been developed.
In recent years, software packages for multistate modeling have become available. R is the language of choice and the Comprehensive R Archive Network (CRAN) is the main repository. The packages are free and the source code is available in CRAN. The packages include survival, eha, Epi, mvna, etm, mstate, msm, Biograph, MicSim andTraMineR. For a recent review, see Willekens and Putter (2014) Software for multistate analysis, Demographic Research 31(14):381-420, and the CRAN Task View on Survival Analysis In 2011, the Journal of Statistical Software published a special issue on multistate modeling (H. Putter, editor). Multistate modeling is an active area of research across disciplines. The research benefits from the current interest in prognostic modeling (of outcomes of health conditions and behavior/lifestyle) and predictive analytics.
https://ir.lib.uwo.ca/pclc_conf/2015/Preconference/1