Publications
publications by categories in reversed chronological order
2024
- Modeling Disability-Free Life Expectancy With Duration Dependence: A Research Note on the Bias in the Markov AssumptionTianyu Shen, and James O’DonnellDemography, Dec 2024. Code can be found here.
Demographic studies on healthy life expectancy often rely on the Markov assumption, which fails to consider the duration of exposure to risk. To address this limitation, models like the duration-dependent multistate life table (DDMSLT) have been developed. However, these models cannot be directly applied to left-censored survey data, as they require knowledge of the time spent in the initial state, which is rarely known because of survey design. This research note presents a flexible approach for utilizing this type of survey data within the DDMSLT framework to estimate multistate life expectancies. The approach involves partially dropping left-censored observations and truncating the duration length after which duration dependence is assumed to be minimal. Utilizing the U.S. Health and Retirement Study, we apply this approach to compute disability-free/healthy life expectancy (HLE) among older adults in the United States and compare duration-dependent models to the typical multistate model with the Markov assumption. Findings suggest that while duration dependence is present in transition probabilities, its effect on HLE is averaged out. As a result, the bias in this case is minimal, and the Markov assumption provides a plausible and parsimonious estimate of HLE.
- Dynamics of Health Expectancy: An Introduction to the Multiple Multistate Method (MMM)Tianyu Shen, Collin F. Payne, and Maria JahromiSociological Methods & Research, Sep 2024. Code can be found here.
Many studies have compared individual measures of health expectancy across older populations by time-invariant characteristics. However, very few have included time-varying variables when calculating health expectancy. Even among older adults, socioeconomic and demographic characteristics are likely to change over the life course, and these changes may have substantial implications for health outcomes. This paper proposes a multiple multistate method (MMM) that situates the multistate model within the broader family of vector autoregressive models. Our approach allows the incorporation of the coevolution of multiple life course factors and provides a flexible yet simple way to model two or more time-varying variables with the multistate model. We demonstrate the MMM in two empirical applications, showing the flexibility of the approach to explore health expectancies with complex state spaces.
- Forecasting school enrollments in the Australian Capital TerritoryTianyu Shen, James Raymer, and Caroline HendyJournal of the Royal Statistical Society Series A: Statistics in Society, Oct 2024. Code can be found here.
School enrollment forecasts are vital for effective planning. This study introduces a probabilistic multiregional population projection model, which accounts for different components, including preschool entries, migration, grade progression, and graduations. Using different distributions with a cohort component projection and Monte Carlo simulation, this paper forecasts student enrollments for each school and academic year level in the Australian Capital Territory based on the annual record-level administrative data. In the in-sample validation tests, the model’s overall performance is robust, and the probabilistic design offers reliable prediction intervals reflecting the variation in observed values. The paper ends with a discussion on the importance of prediction intervals for informing school planning.
- Disability and its impact on life expectancy: heterogeneity across Mexican statesEmerson Augusto Baptista, Tianyu Shen, and Vladimir Canudas-RomoBMC Public Health, Oct 2024.
Background: The percentage of the world’s population with disabilities is estimated to be 16%, although its distribution and intensity varies within nations. We aim to disentangle the degree and types of disabilities, estimate the years spent with more severe disabilities, and analyze their distribution across states and between sexes in Mexico. Methods: The Mexican Census of 2020 includes information on disabilities, which allows the study of its national distribution. We used life tables and the Sullivan method to calculate the number of years spent with disability (NYSD) and its percentage with respect to life expectancy for each state and each sex. Results: In Mexico, the population with disabilities is estimated to be 16.5%. Of this total, 69% have milder disabilities, while the remaining 31% have more severe disabilities. At age eighteen, there is a higher NYSD from more severe disabilities for females with 5.67 years (95% CI 5.66 to 5.69) as opposed to males with 3.66 years (95% CI 3.65 to 3.67). Across states, a more homogeneous distribution with lower NYSD is observed for men (between 2.44 and 5.69 years) than for women (4.14 and 8.08 years). A north-south division can also be observed, with particularly notorious disadvantages among coastal states, which is more distinctive among women. Conclusions: This study shows that comparing the number of years spent with disability and the total life expectancy between subpopulations is essential for monitoring the well-being of aging populations, guiding policy decisions, and promoting a society that values and supports all individuals, regardless of their abilities.
- National Population Growth Rate, Its Components, and Subnational Contributions: A Research NoteVladimir Canudas-Romo, Tianyu Shen, and Collin F. PayneDemography, Jun 2024.
A population’s current growth rate is determined jointly by changes in fertility, mortality, and migration. This overall growth rate is also the average of age-specific growth rates, which can be decomposed into the result of historical changes in fertility, mortality, and migration. However, doing so requires more than 100 years of historical data, meaning that such analyses are possible only in a select few populations. In this research note, we propose an adapted version of the variable-r model to measure contributions to the population growth rate for countries with shorter demographic series. In addition, we extend this model to explore the contribution of subnational changes to the national population growth rate. Our results demonstrate that the age-specific growth rates obtained from short historical series, say 25 years, closely match those of the longer series. These abbreviated age-specific growth rates closely resemble the growth rate at birth of their respective cohorts, which is the major determinant of population growth, except at older ages where mortality becomes the main explanatory element. Exploring subnational populations, we find considerable heterogeneity in the age profile of the components of growth and find that the most populous regions tend to have an outsized impact on national-level growth.
2023
- Modelling the Age and Sex Profiles of Net International MigrationJames Raymer, Qing Guan, Tianyu Shen, Sara Hertog, and Patrick GerlandDec 2023.
This technical paper proposes a methodology for inferring the age and sex profiles of net migration. This approach enhances the capability of the Population Division of the United Nations Department of Economic and Social Affairs (United Nations Population Division) to estimate and project populations for the World Population Prospects (WPP). The age and sex profiles of net migration are crucial inputs to demographic accounting models for population estimation and projection. However, most countries in the world do not directly measure migration, and residual estimation methods for inferring patterns have proven inadequate owing to errors in population measures, births, and deaths. Recognizing that net migration lacks consistent patterns across different ages and sexes, this study introduces a novel strategy: estimating immigration and emigration flows by age and sex—categories that demonstrate regular patterns–and using differences from these flows to estimate net international migration by age and sex. Empirical validations using data from Sweden and the Republic of Korea have yielded promising results, prompting the extension of the method to estimate age-and sex-specific net migration patterns for countries lacking migration data.
- Decomposition of Differentials in Health Expectancies From Multistate Life Tables: A Research NoteTianyu Shen, Tim Riffe, Collin F. Payne, and Vladimir Canudas-RomoDemography, Nov 2023. Code can be found here.
Multistate modeling is a commonly used method to compute healthy life expectancy. However, there is currently no analytical method to decompose the components of differentials in summary measures calculated from multistate models. In this research note, we propose a derivative-based method to decompose the differentials in population-based health expectancies estimated via a multistate model into two main components: the proportion resulting from differences in initial health structure and the proportion resulting from differences in health transitions. We illustrate the method using data on activities of daily living from the U.S. Health and Retirement Study to decompose the sex differential in disability-free life expectancy (HLE) among older Americans. Our results suggest that the sex gap in HLE results primarily from differences in transition rates between disability states rather than from the initial health distribution of female and male populations. The methods introduced here will enable researchers, including those working in fields other than health, to decompose the relative contribution of initial population structure and transition probabilities to differences in state-specific life expectancies from multistate models.
- Disability and morbidity among US birth cohorts, 1998–2018: A multidimensional test of dynamic equilibrium theoryTianyu Shen, and Collin PayneSSM - Population Health, Dec 2023.
A substantial body of prior research has explored patterns of disability-free and morbidity-free life expectancy among older populations. However, these distinct facets of later-life health are almost always studied in isolation, even though they are very likely to be related. Using data from the US Health and Retirement Study and a multistate life table approach, this paper explores the interactions between disability, morbidity, and mortality by sex and education among four successive US birth cohorts, born from 1914 to 1923 to 1944–1953 and compared in the periods 1998–2008 and 2008–2018. We find little compression of disability but a marked expansion of morbidity across cohorts. However, disability-free life expectancy (DFLE) among those living with chronic morbidities has increased, even though at the population-level DFLE is largely unchanged. Broadly, these patterns suggest that successive cohorts of older populations in the US are experiencing a dynamic equilibrium, where the link between chronic morbidities and disability has weakened over successive cohorts. Investigating patterns by educational attainment, we find marked disparities where the least educated individuals not only live significantly fewer years free of disabilities or chronic morbidities but also have experienced an expansion in morbidity and disability. Our findings suggest that the future trajectory of disability-free life expectancy in the US is increasingly contingent on efforts to improve disease management and control the severe consequences of chronic morbidities.
- The estimation of age and sex profiles for international migration amongst countries in the Asia-Pacific regionTianyu Shen, James Raymer, Guan Qing, and Arkadiusz WiśniowskiPopulation, Space and Place, Oct 2023. Interactive results can be found here and full results can be downloaded here.
Information on the age and sex patterns of international migration occurring amongst countries in the Asia-Pacific region is important for researchers and policymakers to understand the contributions of migration towards demographic, economic and social changes. However, despite its importance, statistics on the age and sex patterns are largely absent and, where they do exist, there are considerable differences in measurement and data collection procedures. In this study, we develop a methodology for indirectly estimating the age and sex patterns of international migration for 53 populations in the Asia-Pacific region and four macroworld regions. Multiplicative component models are used to systematically combine information from multiple data sources and infer missing data, resulting in complete and harmonised tables of migration by origin, destination, age, and sex from 2000 to 2019. The estimates are presented and assessed with some discussions on how they may be applied and further improved.
- The contribution of survival to changes in the net reproduction rateTianyu Shen, Ester Lazzari, and Vladimir Canudas-RomoPopulation Studies, Mar 2023. Interactive results can be found here.
The net reproduction rate (NRR) is an alternative fertility measure to the more common total fertility rate (TFR) and accounts for the mortality context of the population studied. This study is the first to compare NRR trends in high- and low-income countries and to decompose NRR changes over time into fertility and survival components. The results show that changes in the NRR have been driven mostly by changes in fertility. Yet improvements in survival have also played an important role in explaining changes in the NRR over the last century and represent a substantial component of change in some low-income countries today. Furthermore, the decomposition of the survival component by age indicates that the survival effect on population reproduction is concentrated mostly in infancy, although the HIV/AIDS epidemic altered this age profile in some populations. The findings highlight the importance of mortality’s effect on reproduction in specific periods and contexts.
2022
- The Components of Change in Population Growth RatesVladimir Canudas-Romo, Tianyu Shen, and Collin F. PayneDemography, Apr 2022. Interactive results can be found here.
The demographic balance equation relates the population growth rate with crude rates of fertility, mortality, and net migration. All these rates refer to changes occurring between two time points, say, t and t + h. However, this fundamental balance equation overlooks the contribution of historical fertility, mortality, and migration in explaining these population counts. Because of this, the balance equation only partially explains a change in growth rate between time t and t + h as it does not include the contribution of historical population trends in shaping the population at time t. The overall population growth rate can also be expressed as the weighted average of age-specific growth rates. In this article, we develop a method to decompose the historical drivers of current population growth by recursively employing the variable-r method on the population’s average age-specific growth rates. We illustrate our method by identifying the unique contributions of survival progress, migration change, and fertility decline for current population growth in Denmark, England and Wales, France, and the United States. Our results show that survival progress is mainly having an effect on population growth at older ages, although accounting for indirect historical effects illuminates additional contributions at younger ages. Migration is particularly important in Denmark and England and Wales. Finally, we find that across all populations studied, historical fertility decline plays the largest role in shaping recent reductions in population growth rates.
- Estimating international migration flows for the Asia-Pacific region: Application of a generation–distribution modelJames Raymer, Qing Guan, Tianyu Shen, Arkadiusz Wiśniowski, and Juliet PietschMigration Studies, Dec 2022. Code and full results can be found here.
Flows of international migration are needed in the Asia-Pacific region to understand the patterns and corresponding effects on demographic, social, and economic change across sending and receiving countries. A major challenge to this understanding is that nearly all of the countries in this region do not gather or produce statistics on flows of international migration. The only information that are widely available represent immigrant population stocks measured at specific points in time—but these represent poor proxies for annual movements. In this paper, we present a methodology for indirectly estimating annual flows of international migration amongst 53 populations in the Asia-Pacific region and four macro world regions from 2000 to 2019 using a generation–distribution framework. The estimates suggest that 27–31 million persons from the Asia-Pacific region have changed their countries of usual residence during each year in the study. Southern Asia is estimated to have had the largest inflows and outflows, whilst intra-regional migration and return migration were highest in Eastern, Southern, and South-Eastern Asia. India, China, and Indonesia were estimated to have had the largest emigration flows and net migration losses. As a first attempt to estimate international migration flows in the Asia-Pacific region, this paper provides a basis for understanding the dynamics and complexity of the large-scale migration occurring in the region.
2021
- The role of reductions in old-age mortality in old-age population growthVladimir Canudas-Romo, Tianyu Shen, and Collin PayneDemographic Research, Dec 2021.
BACKGROUND: The variable-r model provides demographers with a way to explore the contributions of demographic components (fertility, mortality, migration) to changes in populations’ age structures. However, traditional variable-r methods require extremely long mortality series to explore growth at oldest-old ages. OBJECTIVE: Our goal is to disentangle the old-age growth rate into two main components: the growth rate at some younger age, and reductions in mortality between the younger and older ages. METHODS: We focus on an adaptation of the variable-r model that can use shorter mortality series to explore population growth between two ages. RESULTS: Using data from the Human Mortality Database, we explore how these two components are driving the growth rate of 100-year-olds. Observed growth of those reaching age 100 results primarily from the high growth rates when those cohorts were 80-year-olds, and from time reductions in cohort mortality between ages 80 and 100. However, the latter component behaves differently across populations, with some countries experiencing recent slowdowns in cohort mortality declines or increases in mortality between ages 80 and 100. CONCLUSIONS: We find great diversity in the level of old-age mortality improvements across populations, and heterogeneity in the drivers of these improvements. Our findings highlight the need to closely monitor the underlying reasons for the changes in old-age mortality across populations and time. CONTRIBUTION: We present illustrations of the use of the variable-r method to monitor demographic change in an online interactive application, estimated even when only short historical series of demographic data are available.
