Volume 58 - Number 1 - Spring 1992

N. Methods of Research and Analysis Including Models

Studies concerned with demographic methods and with methods from other disciplines that have been applied to demographic data as a whole. Includes mathematical demography and studies on methods of estimation and indirect estimation. Methodological studies and models concerned with one demographic variable, such as migration, are coded under the category concerned with that topic and cross-referenced to this heading. Studies on models used to investigate relationships between demographic variables and for the analysis of empirical data are also coded under this heading.

58:10709 Barr, Thomas H. Approximation for age-structured population models using projection methods. Computers and Mathematics with Applications, Vol. 21, No. 5, 1991. 17-40 pp. Exeter, England. In Eng.
"In this paper we present a class of rapidly convergent numerical schemes to solve the Sharpe-Lotka model equation from age-dependent population dynamics. This work is based on spline approximation techniques described by Banks and Kappel for functional differential equations. We provide a motivation for and description of a generalized problem which under suitable conditions is equivalent to the Sharpe-Lotka problem, we describe approximation in general which exploits the Hilbert space structure in which the generalized problem is set, and for a particular space of approximating functions we obtain estimates on the rates of convergence."
Correspondence: T. H. Barr, Rhodes College, Department of Mathematics and Computer Science, Memphis, TN 38112-1690. Location: Princeton University Library (ST).

58:10710 Becker, Nikolaus; Rittgen, Werner. Fitting cancer mortality data with cumulative damage models. Mathematical Biosciences, Vol. 108, No. 1, Feb 1992. 57-73 pp. New York, New York. In Eng.
"Cumulative damage models conceive the epidemiologically observed aspects of carcinogenesis as some kind of total balance over a complex biological process and suggest that this total balance might behave as a wear-and-tear process. The essential concepts of this mechanistic model are exposure to a carcinogenically damaging environment and resistance of a host system against those damages. Intensity of exposure and magnitude of host resistance are the parameters to be assessed. The paper describes (1) the statistical methods for fitting this model to birth cohort data; (2) for which cancer sites the model provides acceptable fits and for which it does not; and (3) how model extensions provide improvements in the goodness of fit....The data used for this investigation are the [Federal Republic of Germany] statistics for causes of death that are available for the period 1952-1985."
Correspondence: N. Becker, German Cancer Research Center, Department of Epidemiology, Heidelberg, Germany. Location: Princeton University Library (SM).

58:10711 Blythe, S. P. Heterogeneous sexual mixing in populations with arbitrarily connected multiple groups. Mathematical Population Studies, Vol. 3, No. 3, 1992. 173-88, 227 pp. Reading, England. In Eng. with sum. in Fre.
"A simple procedure for constructing [social/sexual] mixing models for arbitrarily classified (e.g. by sex, age, geographical location, sexual preference) populations is outlined, including a scheme for finding the number of independent mixing parameters required, and a simple (linear algebra) means for finding the values of the dependent mixing parameters. Various worked examples are presented, including the two-sex problem and structured and selective mixing." The use of the models for analyzing mixing structures for AIDS transmission is assessed.
Correspondence: S. P. Blythe, University of Strathclyde, Department of Statistics and Modeling Science, Glasgow G1 1XH, Scotland. Location: Princeton University Library (SPR).

58:10712 Blythe, Stephen P.; Castillo-Chavez, Carlos; Casella, George. Empirical methods for the estimation of the mixing probabilities for socially structured populations from a single survey sample. Mathematical Population Studies, Vol. 3, No. 3, 1992. 199-225, 227 pp. Reading, England. In Eng. with sum. in Fre.
"The role of variability of sexual behavior in the transmission dynamics of HIV and AIDS has been illustrated, through the use of mathematical models, by several investigators....In this paper we describe some practical methods for estimating the deviations from random mixing from a single survey sample....We include a description of the role of the estimated mixing probabilities in models for the spread of HIV, a discussion of alternatives and possible extensions of the methods described in this article, and an outline of future directions of research."
Correspondence: C. Castillo-Chavez, Cornell University, Biometrics Unit/Center for Applied Mathematics, Ithaca, NY 14853. Location: Princeton University Library (SPR).

58:10713 Busenberg, Stavros; Castillo-Chavez, Carlos. A general solution of the problem of mixing of subpopulations, and its application to risk- and age-structured epidemic models for the spread of AIDS. Population and Development Program: 1990 Working Paper Series, No. 2.04, 1990. 37 pp. Cornell University, Department of Rural Sociology: Ithaca, New York. In Eng.
This study is concerned with the problem of mixing in the study of sexually transmitted diseases, with particular reference to the spread of AIDS. "In this paper we have extended [a] mixing framework...through the incorporation of age-structure, and have found the general solution to the mixing problem for a sexually-active homosexual population. This general solution, as well as a number of other results, are new even in the simpler context where there is no age-dependence. We have clarified the role of proportionate mixing by showing that it is the only separable solution, and have formulated a general epidemic model for a single, age-dependent, sexually-active homosexual population with distributed activity levels. An explicit expression for the reproductive number for the special case of proportionate mixing has been determined."
Correspondence: Cornell University, Department of Rural Sociology, Population and Development Program, 134 Warren Hall, Ithaca, NY 14853-7801. Location: University of Pennsylvania, Demography Library, Philadelphia, PA.

58:10714 Casetti, Emilio. Bayesian regression and the expansion method. Geographical Analysis, Vol. 24, No. 1, Jan 1992. 58-74 pp. Columbus, Ohio. In Eng.
"This paper shows that within a Bayesian frame of reference the propensity to regard a model as invariant can be formalized into a prior probability density function of the parameters of an expanded formulation in which this model is encapsulated. The Bayesian approach allows to bring to the surface the implications of alternative levels of commitment to invariance assumptions on the results of empirical analyses, and can quantify the comparative strength of alternative drift specifications. The 'expanded' Bayesian regressions are demonstrated by an example focusing upon the effectiveness of family planning [policies]....[It is found that] the primal relationship between fertility decline and level of modernization drifts with level of family-planning effort; and the dual relationship between fertility decline and family-planning effort drifts in the modernization space. The implication of this drift is that the primal relation between fertility decline and modernization...is different depending upon the intensity of the family-planning effort."
Correspondence: E. Casetti, Ohio State University, Department of Geography, 190 North Oval Mall, Columbus, OH 43210. Location: Princeton University Library (UES).

58:10715 Chung, Robert E. Cycles in the two-sex problem: an investigation of a nonlinear demographic model. Pub. Order No. DA9126513. 1990. 101 pp. University Microfilms International: Ann Arbor, Michigan. In Eng.
This work was prepared as a doctoral dissertation at the University of California at Berkeley.
Correspondence: University Microfilms International, 300 North Zeeb Road, Ann Arbor, MI 48106-1346. Source: Dissertation Abstracts International, A: Humanities and Social Sciences 52(4).

58:10716 Cressie, Noel. Smoothing regional maps using empirical Bayes predictors. Geographical Analysis, Vol. 24, No. 1, Jan 1992. 75-95 pp. Columbus, Ohio. In Eng.
The author discusses smoothing regional maps using empirical Bayes predictors. "Section 2 develops the general Bayesian framework....Section 3 presents estimation procedures for model and prior parameters, based on maximum likelihood estimation. Section 4 discusses the important statistical notion of estimating mean squared prediction error accurately. An analysis of sudden infant death syndrome (SIDS) in the one hundred counties of North Carolina, 1974-1978, is presented in section 5; using the theory of sections 2, 3, and 4, the sudden infant death (SID) rate is mapped regionally. Section 6 contains a discussion of some general issues in regional mapping."
Correspondence: N. Cressie, Iowa State University, Department of Statistics, Ames, IA 50011. Location: Princeton University Library (UES).

58:10717 Fratczak, Ewa; Jozwiak, Janina; Paszek, Barbara. The methodology of studies on family and individual life cycles--selected aspects. [Metodyka badan cyklu zycia jednostki i rodziny--wybrane aspekty.] Monografie i Opracowania, No. 341, 1991. 159 pp. Szkola Glowna Handlowa, Instytut Statystyki i Demografii: Warsaw, Poland. In Pol. with sum. in Eng; Rus.
The authors outline the methodology used to study the life cycle of both individuals and families, including event history analysis. The first chapter describes the basic concepts involved and methods of analysis employed. The second chapter reviews the computer programs and packages available for this kind of analysis. Five examples of specific analyses using such programs are given in the third chapter. The fourth and final chapter is devoted to the empirical verification of selected methods of analyzing family and individual life cycles using data from a Polish survey carried out in 1988.
Correspondence: Szkola Glowna Handlowa, Instytut Statystyki i Demografii, Al. Niepodlegosci 162, 02-554 Warsaw, Poland. Location: Princeton University Library (SPR).

58:10718 Jozwiak, Janina; Kotowska, Irena E. Usefulness of demographic modelling. Proceedings of the seminar "Usefulness of Demographic Models," Jadwisin, November, 1989. Monografie i Opracowania, No. 345, 1991. 315 pp. Szkola Glowna Handlowa, Instytut Statystyki i Demografii: Warsaw, Poland. In Eng. with sum. in Rus; Pol.
These are the proceedings of a conference on the usefulness of demographic models, held in Jadwisin, Poland, in November 1989. The 20 papers are organized into five sections. The first two are concerned with the modeling of population dynamics and processes. Other sections deal with demo-economic models, software for demographic analysis, and mathematical models.
Correspondence: Szkola Glowna Handlowa, Instytut Statystyki i Demografii, Al. Niepodlegosci 162, 02-554 Warsaw, Poland. Location: Princeton University Library (SPR).

58:10719 Klissou, Pierre. Multicollinearity in path analysis: an application to demography. [La multicollinearite dans l'analyse des cheminements: un essai d'application a la demographie.] Institut de Demographie Working Paper, No. 160, ISBN 2-87209-155-6. 1991. 26 pp. Universite Catholique de Louvain, Institut de Demographie: Louvain-la-Neuve, Belgium. In Fre.
The author describes the concepts of path analysis and how they have been applied in demographic research, particularly in the study of fertility. Particular attention is given to the problems posed by multicollinearity among independent variables in the study of the effect of changes in such variables on dependent variables.
Correspondence: Universite Catholique de Louvain, Institut de Demographie, Place Montesquieu 1, Boite 17, 1348 Louvain-la-Neuve, Belgium. Location: Princeton University Library (SPR).

58:10720 Mitra, S. Estimates of the parameters of the trajectory of births. Demography India, Vol. 18, No. 1-2, Jan-Dec 1989. 81-93 pp. Delhi, India. In Eng.
"In this paper we have proceeded with the assumption of a simple pattern for the time series generated by the births that can be attributed to the initial population. Assuming that this time series can be described by two intersecting lines with appropriate slopes it has been shown that the coefficients can be expressed in terms of simple functions of these slopes and of other parameters derivable from the distribution of the net maternity rates....An application of the results of this exercise on a hypothetical data set reveals the relative simplicity of this procedure."
Correspondence: S. Mitra, Emory University, Department of Sociology, Atlanta, GA 30322. Location: Princeton University Library (SPR).

58:10721 Rogers, Andrei. Heterogeneity and selection in multistate population analysis. Demography, Vol. 29, No. 1, Feb 1992. 31-8 pp. Washington, D.C. In Eng.
"This paper addresses the selective effects of heterogeneity in interdependent multistate populations experiencing recurrent and reversible events. It demonstrates that in such demographic processes, no a priori conclusion can be made regarding the direction of the selectivity effect, and that the dangers of misspecification from inappropriately defined rates (e.g., prevalence rates) are heightened." Some mortality data for selected developed countries are used as illustrations.
Correspondence: A. Rogers, University of Colorado, Department of Geography and Population Program, Boulder, CO 80309-0484. Location: Princeton University Library (SPR).

58:10722 Trussell, James; Guinnane, Timothy W. Techniques of event history analysis. OPR Working Paper, No. 91-7, Feb 1991. 28 pp. Princeton University, Office of Population Research [OPR]: Princeton, New Jersey. In Eng.
"In this paper, we explicate the logic of failure time models in terms that will be familiar to historical demographers who are conversant with standard statistical techniques such as logistic regression....We...aim to provide practical guidance on the strengths and weaknesses of alternative models and of standard statistical software. In the first section, we provide a brief introduction to event history analysis. In the second section, we consider the generalization of life-table methodology to accommodate covariates. In the third section, we briefly describe several alternative specifications of failure time models....In the fourth section, we discuss the issues of analyses of concurrent processes, noninformative censoring, unobserved heterogeneity, and goodness of fit. We close with an illustrative analysis of remarriage in Germany that demonstrates the power of these new techniques."
Correspondence: Princeton University, Office of Population Research, 21 Prospect Avenue, Princeton, NJ 08544-2091. Location: Princeton University Library (SPR).

58:10723 Wu, Lawrence L. Modeling the shape of the first marriage rate: pooling, statistical power, and nonproportionality. CDE Working Paper, No. 91-17, May 1991. 22 pp. University of Wisconsin, Center for Demography and Ecology: Madison, Wisconsin. In Eng.
"Social scientists using hazard regression methods typically estimate separate models by race and ethnicity using a proportional hazard specification. This research note examines [this practice] for the age-specific rate of first marriage using data from the June 1980 [U.S.] Current Population Survey....The results suggest accepting more parsimonious models that pool across race and ethnicity; moreover, pooling across birth cohort or highest grade completed degrades fit more than pooling across race and ethnicity. The results also strongly reject proportional hazard specifications in favor of nonproportional alternatives. A key to these findings is the use of a suitably flexible model for a time-dependent, nonproportional hazard, which underscores the importance of correctly modeling the shape of the rate."
Correspondence: University of Wisconsin, Center for Demography and Ecology, 4412 Social Science Building, 1180 Observatory Drive, Madison, WI 53706-1393. Location: Princeton University Library (SPR).

58:10724 Yadava, K. N. S.; Kumar, U.; Kushwaha, S. N. S. Population projection up to the lower limit of reproductive period under gradual change in fertility schedule: some alternative models. Demography India, Vol. 18, No. 1-2, Jan-Dec 1989. 73-80 pp. Delhi, India. In Eng.
The authors present models designed to project trends in age distribution for stable and stationary populations. Data from the Rural Development and Population Growth Survey undertaken in Uttar Pradesh, India, in 1978 are used to illustrate the modeling techniques.
Correspondence: K. N. S. Yadava, Banaras Hindu University, Department of Statistics, Varanasi 221 005, India. Location: Princeton University Library (SPR).

Copyright © 1992-1996, Office of Population Research, Princeton University.