Monday, April 24, 2017

wishlist: reports to hear

55-3 Probabilistic Population Projections for Countries With Generalized HIV/AIDS Epidemics

Thursday, April 27, 2017: 1:40 PM
Salon A-5 (Hilton Chicago)
David J. Sharrow University of Washington, Seattle, WA
Jessica Godwin Department of Statistics, University of Washington, Seattle, WA
Yanjun He University of Washington, Seattle, WA
Sam Clark The Ohio State University, Columbus, OH
Adrian Raftery Statistics and Sociology, University of Washington, Seattle, WA
The UN issues probabilistic population projections for all countries to 2100 by simulating future levels of total fertility and life expectancy from Bayesian hierarchical models, and combining the results using the cohort-component projection method. However, projections for the 40 countries with generalized HIV/AIDS epidemics used Spectrum/EPP model, a complex 15-compartment model designed for short-term projections of quantities relevant to policy for the epidemic. We propose a simpler approach that is more compatible with the existing UN methodology for other countries. Changes in life expectancy are projected probabilistically using a time series regression model on current life expectancy, HIV prevalence and ART coverage. These are converted to age- and sex-specific mortality rates using a new family of model life tables designed for countries with HIV/AIDS epidemics which are input to the cohort-component method. The method performed well in out-of-sample cross-validation, and gives similar results to Spectrum/EPP projections in the short run.

25-2 Probabilistic Projection of Subnational Total Fertility Rates

Thursday, April 27, 2017: 10:35 AM
Joliet Room (Hilton Chicago)
Hana Sevcikova University of Washington, Seattle, WA
Adrian Raftery Statistics and Sociology, University of Washington, Seattle, WA
Patrick Gerland United Nations Population Division, New York, NY
We consider the problem of probabilistic projection of the total fertility
rate (TFR) for subnational units. We seek a method that is consistent
with the UN's recently adopted Bayesian method for probabilistic TFR
projections for all countries, and works well for all countries.
We assess various possible methods using subnational TFR data for 47 countries.
We find that the method that performs best in terms of out-of-sample
predictive performance and also in terms of reproducing the within-country
correlation in TFR is a method that scales the national trajectory by
a subregion-specific scale factor that is allowed to vary slowly over time.
This supports the hypothesis of Watkins (1990, 1991) that within-country
TFR converges over time in response to country-specific factors,
and extends the Watkins hypothesis to the last 50 years and to a much
wider range of countries around the world.

74-4 Comparing Artificial Neural Network and Cohort-Component Models for Population Forecasts

Thursday, April 27, 2017: 3:45 PM
Joliet Room (Hilton Chicago)
Viktoria Riiman Center for Business and Economic Research, University of Alabama, Tuscaloosa, AL
Amalee Wilson Computer and Information Sciences, University of Alabama at Birmingham, Birmingham, AL
Reed M Milewicz Computer and Information Sciences, University of Alabama at Birmingham, Birmingham, AL
Peter Pirkelbauer Computer and Information Sciences, University of Alabama at Birmingham, Birmingham, AL
Artificial neural network (ANN) models are rarely used to forecast population in spite of their growing prominence in other fields. We compare the forecasts generated by ANN long short-term memory models with population projections from traditional cohort-component method (CCM) for counties in Alabama. The evaluation includes forecasts for all 67 counties that offer diversity in terms of population and socioeconomic characteristics. When comparing predicted values with total population counts from the 2010 decennial census, the CCM used by the Center for Business and Economic Research at the University of Alabama in 2001 produced more accurate results than a basic multi-county ANN model. Only when we use single-county models or proxy for a forecaster's experience and personal judgment with potential economic forecasts, results from ANN models improve. The results indicate the significance of forecaster's experience and judgment for CCM and difficulty, but not impossibility of substituting these insights with available data.

P3-45 Immigration Forecasts: A Bayesian Semiparametric Approach for Seasonal Data

Thursday, April 27, 2017
Salon D (Hilton Chicago)
Alice Milivinti University of Geneva, Geneva, Switzerland
Giacomo Benini University of Geneva, Geneva, Switzerland
Immigration is the main driver of population change in developed countries. Nevertheless, a
common sense around its unpredictability has left migration forecasts relatively underdeveloped
vis-à-vis other types of population projections. Using high-frequency data, this paper proposes
a new procedure specifically designed to capture seasonal irregularities generated by trend-cycle
interactions. The resulting estimation strategy identifies a class of structural parameters which
reflect the non-constant nature of the immigration flows while introducing prior information
derived from subjective expectations. The outcome is an efficient symbiosis between Bayesian
probability and semiparametric flexibility, which outperforms in terms of forecast accuracy a
set of alternatives

23-1 Socially Embedded Preferences, Environmental Externalities, and Reproductive Rights

Thursday, April 27, 2017: 8:30 AM
Continental B (Hilton Chicago)
Aisha Dasgupta United Nations Population Division, New York, NY
Partha Dasgupta Department of Economics, Cambridge University, Cambridge, United Kingdom
We review a class of adverse environmental externalities that accompany consumption and procreation. We also identify externalities that are traceable to socially embedded preferences for family size. Those preference structures can give rise to a heightened demand for children, exacerbating the environmental externalities households impose on future generations. Our analysis exposes weaknesses in basing family planning programmes entirely on individuals' reproductive rights. We use ecological data to obtain a feel for the size of global environmental externalities. We estimate the world population the biosphere can support at the standard of living enjoyed in high middle-income countries. Today's and future global population projections far exceed our estimate, implying that the UN's Sustainable Development Goals are unsustainable. We conclude that family planning has been undervalued by national governments and international agencies. Our purpose is to pose questions that continue to be neglected in the development literature. We do not make policy recommendations.

P3-84 Brass Relational Logit Model and Life Table Construction

Thursday, April 27, 2017
Salon D (Hilton Chicago)
Lei LU Freelancer, Beijing, China
The Brass Relational Logit Model (BRLM) has been one of the most commonly used methods for (model) life table construction and mortality estimates. BRLM can be used as a statistical model, by which the two parameters (alpha and beta) can be estimated from statistical data. In the meantime, the logit model can also be used as a two-parameter analytic model, by which the relationship between two life tables can be studied mathematically.
In this paper, the author discusses three issues related to (model) life table construction with the Brass relational logit model, i.e. (i) some basic analytic properties of the Brass relational logit model, (ii) the analytic relationships between the alpha parameter and various life table indicators, and (iii) the analytic effects of the death distribution factor (i.e. a(x,n) ) on various life table indicators. Some of the analytic relationships/equations developed in this paper can be used for mortality estimation and projection purposes.

55 HIV/AIDS: Methods and Impact

Thursday, April 27, 2017: 1:00 PM-2:30 PM
Salon A-5 (Hilton Chicago)
Session Chair: Mark E. McGovern

26 Demographic Applications in the Private and Public Sectors

Thursday, April 27, 2017: 10:15 AM-11:45 AM
Astoria Room (Hilton Chicago)
Discussant: Beth Jarosz
Session Chair: Mike Cline

132 Statistical Methods in Population Research

Friday, April 28, 2017: 8:30 AM-10:00 AM
Williford C (Hilton Chicago)
Discussant: Scott M. Lynch
Session Chair: Shooshan Danagoulian

146 Health and Mortality Projections

Friday, April 28, 2017: 10:15 AM-11:45 AM
Salon A-1 (Hilton Chicago)
Discussant: Simona Bignami
Session Chair: Vinod Mishra
Program in general

No comments:

Post a Comment