No full-text available
To read the full-text of this research,
you can request a copy directly from the author.
Abstract
Population projections generally use the component method which consists in adapting sample size by age and sex from mortality, fertility and migration hypotheses. This exercise is not always aimed a providing projection data, but may also serve to analyze factors changing in the population. Comparatively long-term demographic forecasts, particularly for mortality, can be made due to the major inertia of demographic phenomena. However, the quality of such predictions depends on the quality of knowledge of past evolution in the demographic factors and on the capacity to forecast their future evolution.
ResearchGate has not been able to resolve any citations for this publication.
Projections of births and deaths can be based on different assumptions : the situation becomes even more complex when migration is included. In general, emigration rates have been calculated and applied to the populations of countries of origin. The end result is a Markov process, with well known properties, and which ultimately leads to stable regional populations. However, it has been shown in many studies that such rates are not very satisfactory to describe migration flows between different areas. Indices which include the populations of both the region of origin and the region of arrival are more satisfactory, both from the theoretical and the practical points of view. The use of such indices over a finite period (between two censuses, for instance), leads to results very different from those obtained by using rates of emigration. In particular, populations do not tend towards stability, and some may even become extinct, or even negative. These difficulties are the results of using a discrete-time model ; we therefore investigate what happens if continuous-time models are used. The difficulties caused by discrete- time models can be overcome. In all
circumstances, there is a unique solution for a given initial population and hazards of moving. A number
of continuous-time and discrete-time models are compared. However, the use of continuous-time
models leads to problems of its own. e.g. how to estimate a hazard function which would enable us to
find the total number of migrants during an intercensal period. Although these methods do not result in
stable populations, they make it possible to disentangle the interactions that exist between populations
of different regions and to show the consequences of this distribution. The methods should be
generalized to populations to apply to populations disaggregated into sub-groups, e.g. age groups,
occupational groups, or marital status groups.
Microsimulation differs from traditional macrosimulation in using a sample rather than the total population, in operating at the level of individual data rather than aggregated data, and in being based on repeated random experiments rather than average numbers. Here are presented the circumstances in which microsimulation san be of greater value than the more conventional methods. It is particularly relevant when the results of the process being studied are complex whereas the forces driving it are simple. A particular problem in microsimulation results from the fact that the projections are subject to random variation. Various sources of random variations are examined but the most important is the one we refer to as specification randomness: the more explanatory variables are included in the model, the greater the degree of random variation affecting the output of the model. After a brief survey of the microsimulation models which exist in demography, a number of the essential characteristics of microsimulation are illustrated using the KINSIM model for projecting the future size and structure of kinship networks.
PIP:
Population projections are presented for France up to the year 2050. Three alternative hypotheses concerning fertility are included in the projections. The projections indicate that the population will continue to grow in size up to the year 2020. At that time, the population over age 60 will be larger both absolutely and proportionately than the population under age 20. (SUMMARY IN ENG AND GER AND SPA)