Gender-Structured Population Modeling

From MaRDI portal





In this book the authors give a brief historical perspective of deterministic modeling of human populations and then focus on pair formation (marriage) and two-sex models. They describe several models, derive theoretical results that show these are well posed, and try to elucidate which marriage function might make a better choice for a particular population -- in one example, that of the United States. The authors describe numerical methods to approximate the solution of the differential models, which is equivalent to creating discrete simulators. They present comparisons of simulation results with actual demographic data in the hope that they will help the believer better understand some of the difficulties concerning the availability of data and demographic modeling, and that they might convince the skeptic that mathematical demography does provide reasonable qualitative and quantitative estimates. Sharpe and Lotka's model describes the dynamics of the births and is based on an integral equation of Volterra type. In analyzing this model Lotka also realized that, though the age density of the population varies in time, it cannot take completely arbitrary values and it should eventually stabilize. Though he was not able to rigorously establish this fact, he offered geometric arguments in support of the thesis. The model is essentially a single-sex model derived under the assumption that the other sex has consistent behavior. This model is discussed here because it forms the cornerstone of contemporary mathematical demography. Leslie considered the mathematical properties of the model more thoroughly as well as the resulting convergence to a stable age distribution. Although Leslie presented the matrix approach to stable population theory, his results were ignored by both ecologists and demographers for the following twenty years. Interest in Leslie models in demography has aroused after Keyfitz and Goodman demonstrated their equivalence to the existing continuous theory. Then authors discuss the stable theory. The stable population theory develops a machinery for the investigation of the longterm behavior of the population. However, the models of the stable theory are built and are valid under a set of assumptions that in reality are frequently not true. As a central concept in demography, the stable population theory has always attracted the interest of scientists. One of the most natural extensions of the one-sex theory is to include both sexes. The one-sex stable population theory is a fundamental theoretical tool in mathematical demography, but it cannot answer questions related to the existence and interplay of the sexes. When applied to one of the sexes, it is implicitly assumed that the other sex has a consistent development, but this approach cannot capture well situations in which one of the sexes is scarcer than the other. In human populations such a disproportion can be observed when the so-called marriage squeeze occurs. Thus, a fundamental shortcoming of the classical stable population theory is that it fails to handle populations differentiated by sex. The Fredrickson-Hoppensteadt model is based on the explicit assumption that the population is actually monogamous. Based on the couples dynamics, a couple, though not necessarily legally formed, is, however, the exclusive status that allows reproduction. That is, the model does not allow births outside marriage (or outside a couple). Motivated by their choice, the authors continue with the investigation of the mathematical properties of the model. It is established that the Fredrickson-Hoppensteadt model is well posed in the sense that, given a set of initial conditions, it has a unique nonnegative solution. Also it is established that, under appropriate conditions, the solution is differentiable and satisfies the equations in the classical sense. Knowledge of the mathematical aspects of the model will enable to set up numerical methods, which are presented in the book. In particular, to prove convergence of the finite difference algorithm used as a basis for the simulator, are assumed certain differentiability properties of the solution. The simulations performed show that the Fredrickson-Hoppensteadt model can be successfully used for projecting human populations -- at least for short intervals of time. But the authors consider that ``reality crashes everywhere and that further modeling will be necessary to reach a better description. Interest in age-structured two-sex models is relatively recent, although the models date from the 1970s and 1980s. The early analysis of such models is mostly restricted to special, simpler cases. Most of the results on the well-posedness of age-structured two-sex population models are concerned with existence and uniqueness of solutions in spaces of integrable functions. Inaba considered a two-sex model for human reproduction by first marriage -- that is, newborns are produced only in first-marriage couples -- and establishes the well-posedness of the model and the existence of persistent solutions. Therefore, since the numerical methods require regularity of the solution, it may first approximate a model with an approximating marriage function and then discretize this modified model to provide numerical approximations. If \(E\) is chosen sufficiently small (e.g., smaller than the machine zero of the computer used), the numerical approximations obtained for the modified model will be identical to those obtained for the original one with the harmonic mean marriage function that results from \(e=0\). Furthermore the authors are concerned with the description of a numerical method -- essentially that of Arbogast and Milner -- that is based on the finite difference method of characteristics for the discretization of the differential equations, combined with quadrature rules for the integrals that describe the reproductive process of the population and the age densities of married females and males. The simulation running results conducted on a special sample of USA population shows that there are noticeable differences in the errors leading to the conclusion that, in order to make projections with the two-sex model, one must choose the marriage function that provides the best fit. This is also the reason why Henry's and McFarland's models perform better than traditional marriage functions. The male dominance function is certainly not the best theoretical marriage function but is it not still possible that the fact that it leads to better projections is a consequence of an actual demographic phenomenon? Indeed, if the marriage market obeys market laws, the outcome -- the number of marriages -- should be more heavily influenced by the scarcer sex. And the fact is that after the age of 20 -- which includes most marriagable ages -- females outnumber males. This observation somewhat contradicts the general perception that females have greater impact on marriages and that a marriage function (if not symmetric) should be slightly female dominant. The observed pattern in the behavior of the errors can be used as a criterion for choosing an appropriate marriage function to improve the projective properties of the simulator. In fact, as can be seen from the authors' experiments, the total number of newborn girls and boys and the total number of couples can be obtained with very good accuracy, provided that an appropriate marriage function is used. In fact a new, more complex marriage function can be proposed by assigning for the number of marriages in each bride-groom age category the projection from the marriage function that results in the largest number of marriages.




Cited in
(63)








This page was built for publication: Gender-Structured Population Modeling

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4679155)