A survey of census bureau population projection methods

Long, John F.; McMillen, David Byron

doi:10.1007/bf00138799

Cited by 19 publications

(9 citation statements)

References 51 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Systems of national and regional demographic accounts--incorporating stocks and flows of people among demographic and/or social statuses such as age and occupational categories--also have increasingly been recognized as defining the context of demographic and social forecasting (see, e.g., Long and McMillen, 1987). These accounts lead, for instance, to the population transformation equation: _I,, =-Ov,_~, , OK,_~,, + Bt (2) This equation says that a column vector of population inflows into a fixed geographic region from period t -1 to period t(Ig,_l.,) is equal to the product of an outflow transition proportions matrix from t --1 to t (Op,_l,,) times a column vector of population outflows from t -1 to t(OK,_l.,) plus a column vector of births and net migrants in period (Bt)t. While equation (2) holds trivially in scalar form for total population aggregates, it becomes virtually indispensible analytically when members of the population are classified by demographic and/or social statuses (e.g., age, location of residence, employment status, occupational category, marital status, attitudes, etc.…”

Section: The Limits Of Identities and The Need For Parameterizationsmentioning

confidence: 99%

“…For instance, when the identity is applied to the determination of the expected future size and age composition of a national population, one conventional approach is to treat net migration as a constant, while expected future changes in fertility and mortality rates are made arbitrary functions of time on the basis of historical experience and expert judgment (Keyfitz, 1977:74-80;Long and McMillen, 1987). Various other more formal modeling procedures have been developed in the literature.…”

Section: The Limits Of Identities and The Need For Parameterizationsmentioning

confidence: 99%

“…For instance, the ECESIS regional economic-demographic model described by Long and McMillen (1987) contains over 7000 equations and over 10,000 endogenous and exogenous variables. Some macroeconomic models of the U.S. economy--even without regional or demographic detail--contain over 1000 equations.…”

Section: System Complexity and Scalementioning

confidence: 99%

“…In practice, this condition seems not to have been applied to empirical forecast records, perhaps because it is too simple. There are a number of other factors, such as the deterioration of data quality for highly disaggregated variables (cf., Long and McMillen, 1987) and the statistical instability of parameter estimates, that mitigate against gains in accuracy in large models.…”

Section: Model Scale Complexity and Verificationmentioning

confidence: 99%

See 3 more Smart Citations

Forecasting in the social and natural sciences: An overview and analysis of isomorphisms

Land

Schneider

1987

Climatic Change

View full text Add to dashboard Cite

This article identifies and analyzes several points of similarity in the structure and context of forecasting in the social and natural sciences. These include: the limits of identities or universal laws as a basis for forecasts; the corresponding need for simplifying parametric representations of one or more of the variables that enter into identities; various sources of uncertainty about parameterizations; intrinsic limitations on predictability or forecasting accuracy in large-scale systems; the need for sensitivity analyses of model responses to changes in exogenous variables and/or parametric structures; problems of model linkage; and the social (organizational and political) context of forecasts. Suggestions for future lines of inquiry are made in each case. Several of these are such that they can benefit from a sharing of experience and expertise across disciplinary lines.Forecasting is an estimation or calculation of future events or developments, derived from a model, simple or complex, heuristic or analytic. Forecasting, to put it in Other words, pertains to the formation of expectations about future states or processes of specific historical entities (cf., Duncan, 1969;Henschel, 1976;Schuessler, 1968). Social scientists perhaps are best known and/or notorious for economic and demographic forecasts, whereas natural scientists typically are associated with and/or blamed for weather, climate, earthquake, and ecological forecasts. But each produces forecasts of many other social and natural phenomena, and there are interstitial areas, such as energy, transportation, and technological forecasting to which both have contributed.Forecasts can be made for single variables or multi-dimensional distributions at a specific time, or for time series of such variables or distributions. They can involve elaborate methodological formalism based on universal laws or identities; can be based on simple extrapolation of past performance or human judgment; or can involve some combination of lawful behavior and extrapolation and judgment--so-called "contextual laws." The "future" forecasted can be distant or near. The scope of forecasts can encompass the world's physical/social/economic systems or a single local phenomenon.The uses of forecasts range from their explanatory value in a scientific sense to their fulfullment of a statutory requirement for practical policy uses.Many of the dangers and pitfalls to which forecasts are subject are discussed in this work. These include intrinsic limits on their accuracy; the sources and impacts of errors in forecasts; the dangers of spurious accuracy and of a false sense of precision; the danger that certain forecasts, when taken seriously, can create their own self-fulfilling prophecy; and the potential conflicts of interest among forecasters and/or users of forecasts.Despite the presence of these and other problems in forecasting, there are, of course, many potential benefits. Not least of these is the possibility that forecasts of important variables can be correct! Moreover, eve...

show abstract

Section: The Limits Of Identities and The Need For Parameterizationsmentioning

confidence: 99%

Section: The Limits Of Identities and The Need For Parameterizationsmentioning

confidence: 99%

Section: System Complexity and Scalementioning

confidence: 99%

Section: Model Scale Complexity and Verificationmentioning

confidence: 99%

See 2 more Smart Citations

Forecasting in the social and natural sciences: An overview and analysis of isomorphisms

Land

Schneider

1987

Climatic Change

View full text Add to dashboard Cite

show abstract

“…Most models aim to reflect the average night-time residential population distribution, though LandScan reflects "ambient" population, the average between night-time residential and daytime commuter populations [21]. To estimate gridded population figures beyond the year of the last census; birth, migration, and death rates are used to project new population totals by areal unit [24]. "Bottom-up" gridded population estimates are derived from micro-census population counts in a sample of areas, or from assumptions about the average household size, and have only recently been developed [25,26].…”

Section: Introductionmentioning

confidence: 99%

How Accurate are WorldPop-Global Gridded Population Data at the Cell-Level?: A Simulation Analysis in Urban Namibia

Dr¹,

Leasure²,

Bird³

et al. 2021

Preprint

View full text Add to dashboard Cite

Disaggregated population counts are needed to calculate health, economic, and development indicators in Low- and Middle-Income Countries (LMICs), especially in settings of rapid urbanisation. Censuses are often outdated and inaccurate in LMIC settings, and rarely disaggregated at fine geographic scale. Modelled gridded population datasets derived from census data have become widely used by development researchers and practitioners; however, none of these datasets have been evaluated for accuracy of population estimates at the grid cell-level. This is because the finest-scale population figures generally available to data producers are those input into gridded population models and disaggregated to smaller grid cells (e.g., 100x100m). We simulate a realistic "true" 2016 population in Khomas, Namibia, a majority urban region, and introduce realistic levels of outdatedness (over 15 years) and inaccuracy in slum, non-slum, and rural areas. We then aggregate these simulated realistic populations by census and administrative boundaries (to mimic census data), and generate 32 gridded population datasets that are typical of a LMIC setting using WorldPop-Global's gridded population approach. We evaluate the cell-level accuracy of these simulated WorldPop-Global datasets, using the original "true" population as a reference. In our simulation, we found large cell-level errors, particularly in urban cells, driven by WorldPop-Global's use of average population densities in large areal units to determine cell-level population densities. Age, accuracy, and aggregation of the input data did play a primary role in these errors. We suggest incorporating finer-scale training data into gridded population models generally, and WorldPop-Global in particular (e.g., from simulated populations, routine household surveys, or slum community profiles), and use of new building footprint datasets as a covariate to improve cell-level accuracy of gridded population data. It is important to measure cell-level accuracy of all gridded population datasets, especially if they are to be used for monitoring key development indicators.

show abstract

The impact of forecasting methodology on the accuracy of national population forecasts: Evidence from the Netherlands and Czechoslovakia

Keilman

Kučera

1991

Journal of Forecasting

View full text Add to dashboard Cite

"This study considers the accuracy of national population forecasts of the Netherlands and the Czechoslovak Socialist Republic.... We look at the demographic components employed in each forecast, the procedure to extrapolate fertility and the level at which assumptions for each component are formulated. Errors in total population size, fertility, mortality and foreign migration, and age structure are considered. We discuss trends in errors and methodology since 1950 and compare the situations in the two countries. The findings suggest that methodology has only a very limited impact on the accuracy of national population forecasts."

show abstract

A survey of census bureau population projection methods

Cited by 19 publications

References 51 publications

Forecasting in the social and natural sciences: An overview and analysis of isomorphisms

Forecasting in the social and natural sciences: An overview and analysis of isomorphisms

How Accurate are WorldPop-Global Gridded Population Data at the Cell-Level?: A Simulation Analysis in Urban Namibia

The impact of forecasting methodology on the accuracy of national population forecasts: Evidence from the Netherlands and Czechoslovakia

Contact Info

Product

Resources

About