Improvingaccuracy in multiple regression estimates of population using principles from causal modelling

This paper reports a mildly restricted procedure for using a theoretical causal ordering and principles from path analysis to provide a basis for modifying regression coefficients in order to improve the estimation accuracy of the ratio-correlation method of population estimation. The modification is intended to take into account temporal changes in the structure of variable relationships, a major element in determining the accuracy of post-censal estimates. The modification of coefficients is conservative in that it uses rank-ordering as a basis of change. Empirical results are reported for counties in Washington state that demonstrate the increased accuracy obtained using the proposed procedure.     

Criteria for selecting a suitable method for producing post-2000 county population estimates: A case study of population estimates in Illinois

We compared 2000 county population estimates for Illinois against 2000 census counts. Administrative records (ADREC) and ratio correlation (Ratio-CORR) methods were used to produce two sets of controlled county estimates for 2000; a third set represented an average of the estimates reached using these methods. Another set using the ADREC method was not controlled to any estimate. Also, the 2000 estimates were adjusted for undercount in the 1990 census. We compared performance of these methods with the performance of two naive models: (i) do nothing and (ii) constant growth rate. ADREC estimates were more accurate than estimates from the Ratio-CORR or Average method in terms of Mean Absolute Percent (MAPE) or weighted MAPE. Undercount adjustment in general improved the accuracy of the estimates for all three methods. A top-down or bottom-up approach worked equally well. As a single method, ADREC performed best.     

A multiple regression method for analysing stage-frequency data

Summary A linear regression method that allows survival rates to vary from stage to stage is described for the analysis of stage-frequency data. It has advantages over previously suggested methods since the calculations are not iterative, and it is not necessary to have independent estimates of stage durations, numbers entering stages, or the rate of entry to stage 1. Simulation is proposed to determine standard errors for estimates of population parameters, and to assess the goodness of fit of models.     

Improving the measurement of temporal change in regression models used for county population estimates

The ratio-correlation method of population estimation is shown to contain an inconsistent temporal relationship between the model's empirical structure and its actual application. A simple transformation of the model's variables is provided that eliminates the inconsistency. Two tests of the relative accuracy of the original and transformed models show that the transformed model achieves accuracy levels equal to or higher than the original. In one test, all nine years show a higher degree of accuracy, of which four are statistically significant. Several possible reasons are given for the increased accuracy shown by the transformed model. The transformation, termed the "rate-correlation" model, is recommended as a logical starting point in the examination of coefficient stability and spatial autocorrelation as well as a method for estimating small populations.     

Confidence intervals for postcensal state population estimates

This paper develops a methodology for constructing confidence intervals around postcensal state population estimates. Using regression equations, forecast intervals are derived around the average age-specific death rates over the postcensal estimation period. These results, combined with the number of postcensal deaths and the most current census counts, are translated into confidence intervals for the age structure. Two approaches are offered for constructing total population confidence intervals. One examines a simulated distribution while the other focuses on the mathematical derivation of population means and variances. The methodology is illustrated by deriving statistically defensible confidence intervals around the July 1, 1975 population of Florida.     

A method for combining sample survey data and symptomatic indicators to obtain population estimates for local areas

Past efforts of statistical demographers to compute postcensal population estimates for local units have been hampered by the fact that they have had to rely completely on symptomatic information. In this paper, a new method of postcensal estimation is presented in which the symptomatic information is combined with sample data by means of a regression format. Combining symptomatic information on births, deaths, and school enrollment with sample data from the Current Population Survey, county estimates of population growth were computed by means of the new method for the postcensal period after 1960. These estimates were tested for accuracy by comparison with a set of special censuses which were conducted between 1964 and 1967 in 75 counties. The results of this test are promising, but not conclusive. A more conclusive test is currently underway using 1970 data. While the method has been tested as a means for estimating population growth, it is to be emphasized that it can be used to compute postcensal estimates for any variable for which the necessary symptomatic information and sample data are available.     

Note on the accuracy of local population estimates

Mr. Silcock's article will be of interest to all concerned with local population data. It may be useful to supplement it by a brief account of the fuller examination of the local population estimates made in 1951 by the General Register Office, since this covered all 1472 administrative areas in England and Wales and could be made in more detail than was possible for a private investigator.

Any census, of course, provides information not available, at least in such detail, at other times or from other sources, and also serves as a base from which estimates for succeeding years can be derived. In addition, however, the General Register Office takes the opportunity of a census to try and assess the accuracy of the various types of current population estimates made by the Department. In the case of local administrative areas the comparison of actual and expected populations made after the 1931 Census is discussed in the Text Volume of the Registrar General's Statistical Review for 1930 (pages 100-102).     

Population estimates for Bangladesh: The use of a specific transitional population model

Summary The development of population in Bangladesh was affected by a succession of man-made and natural calamities, such as the Bengal Famine of 1943, refugee movements following the partition of India and Pakistan in 1947, devastating floods and cyclones around 1970, and the military action during the war of liberation. Though there had been a tradition of census taking and vital registration in Bangladesh, as part of the Indian sub-continent, extending for over a century, vital registration was so deficient as to be almost valueless, and there were gross misstatements of age and under-enumeration in the censuses. In the census of 1941, on the other hand, political manoeuvring led to a substantial overcount of the population. In this paper, Bangladesh population trends are studied within the broader framework of the subcontinent, taking account of plausible differentials. A considerable element of uncertainty was introduced into growth trends as a result of variations in the completeness of census-taking and of unrecorded refugee and labour movements across open land borders. In this connection the substantial inflationary bias associated with techniques of population estimation using the dual record system is discussed. The application of stable population models is even less justified in Bangladesh with its history of declining mortality. A transitional age structure model was constructed on the basis of the information available on declining mortality and accelerating growth and the model was made even more specific by modifications which took care of the impact of recent calamities and of unrecorded migration. The population base of the census of 1961 was adjusted in accordance with this model. The local mortality age pattern was used in projecting the population by sex and age groups to the date at which the census was originally due to be taken in 1971, and to the date when it was actually taken in March 1974. The post-1970 calamities and their effect on mortality were ignored. The aggregate estimate of population of 72.9 million in March 1974 is slightly in excess of the census count (by about two per cent) reported provisionally as 71.3 million. The excess in our estimate could be accounted for by the losses due to cyclone and military action.     

Immigration data and national population estimates for the United States

The immigration component in national population estimates is comparatively small, but it is not insignificant and may indeed be an important source of error. Therefore, it warrants the considera-tion of those concerned with population estimates. The paper considers alternative methods for deriving estimates of immigration from the raw data and presents estimates of net immigration from 1950 to 1965. They are developed from estimates previously published by the Bureau of the Census, but they differ at some points where new data have become available or where a review of the data has led to a change in judgment on how best to use them. The paper also presents suggestions on how immigration statistics might be altered for purposes of improving the estimates.Census data may be used to estimate net immigration by three different methods, but upon analysis each method proves to be inadequate. Hence, data based on visas surrendered at the port of entry must be the principal source of immigration estimates. These data have their limitations because (1) they do not cover net arrivals of citizens from abroad and from Puerto Rico, (2) they do not report departures of aliens, and (3) they do not allocate all immigrants to year of entry. Alien registration and passenger data offer possible alternative estimates.The paper attempts to measure unrecorded immigration, discusses how net arrivals of citizens from abroad and from Puerto Rico may be estimated, and how the age, sex, and race of immigrants may be treated.     

The impact of incomplete geocoding on small area population estimates

Small-area population estimates are often made using geocoded address data in conjunction with the housing-unit method. Previous research, however, suggests that these data are subject to systematic incompleteness that biases estimates of race, ethnicity, and other important demographic characteristics. This incompleteness is driven largely by an inability to complete georeference address-based datasets. Given these challenges, small-area demographers need further, and to date largely unavailable, information on the amount of error typically introduced by using incompletely geocoded data to estimate population. More specifically, we argue that applied demographers should like to know if these errors are statistically significant, spatially patterned, or systematically related to specific population characteristics. This paper evaluates the impact of incomplete geocoding on accuracy in small-area population estimates, using a Vintage 2000 set of block-group estimates of the household population for the Albuquerque, NM metro area. Precise estimates of the impact of incomplete geocoding on the accuracy of estimates are made, associations with specific demographic characteristics are considered, and a simple potential remediation based on Horvitz-Thompson theory is presented. The implications of these results for the practice of applied demography are reviewed.     

A note on the use ofregression methods in population estimates

Evidence which has emerged in the past few years indicates that the relative accuracy of population estimates derived from the ratio-correlation method and the difference-correlation method varies from state to state. In assessing the possible reasons why neither technique is uniformly more accurate, attention is focused on the temporal instability of the statistical relationships between symptomatic indicators and population change. The author concludes that further improvement in population estimates based on regression techniques is likely to be limited until demographers derive means of measuring and adjusting for these temporal changes.     

Comparison of microcomputer programs for making population projections: an update

"There are many programs for making population projections now available for use with microcomputers. This article reviews six of approximately 15 microcomputer population projection programs. Each program is compared to a standard set of criteria relating to such items as hardware and software requirements, input data requirements and specification of assumptions, methodology and documentation, and summary output indicators. Numerical results from projections of six test data sets reflecting different assumptions about mortality, fertility, and migration are compared. Qualitative comments are included for describing special features and for making an overall assessment of each program." (SUMMARY IN FRE)     

Comparison of microcomputer programs for making population projections: An update

There are many programs for making population projections now available for use with microcomputers. This article reviews six of approximately 15 microcomputer population projection programs. Each program is compared to a standard set of criteria relating to such items as hardware and software requirements, input data requirements and specification of assumptions, methodology and documentation, and summary output indicators. Numerical results from projections of six test data sets reflecting different assumptions about mortality, fertility, and migration are compared. Qualitative comments are included for describing special features and for making an overall assessment of each program.     

The effect of undercount on the accuracy of small-area population estimates: Implications for the use of administrative data for improving population enumeration

The effects of underenumeration on the accuracy of alternative methods of population estimation have not been sufficiently analyzed. Although the US Bureau of the Census has decided not to adjust either the counts or its estimates for underenumeration in 1990, the extent to which local population estimates may account for underenumeration is of importance both for those who may wish to adjust existing estimates and in anticipation of future census adjustments. This paper examines the accuracy of small-area population estimation methods with and without adjustment. Mean Percent Errors, Mean Absolute Percent Errors, and Mean Percent Absolute Differences between local estimates for 1990 and 1990 adjusted and unadjusted census counts are computed. Population estimates for 1990 made using housing unit, ratio correlation, and component methods are compared for 451 counties and 2,633 places in the states of California, Florida, Texas, and Wisconsin. An analysis of the data for counties shows little indication that local estimates more accurately estimate the adjusted than the unadjusted population counts. The results for places show clear improvements in accuracy for places in Florida and Texas. Implications of the findings for issues related to undercount adjustment and local population estimates are discussed.     

Small-area population forecasting in an urban setting: a spatial regression approach

This study revisits a spatial regression approach for small-area population forecasting that considers not only direct drivers of local area population growth but also neighbour growth and neighbour characteristics. Previous research suggested that the approach does not outperform extrapolation projections, the currently most-often-used small-area population forecasting technique. We argue the reason is that population growth is affected by its influential factors differently in urban, suburban, and rural areas. Therefore, we hypothesize that the spatial regression forecasting approach can perform better in one type of area at a time, where the influential factors’ effects on population growth can be estimated more accurately. This study is focused on census tracts of the city of Milwaukee, USA, to test the performance of the spatial regression approach in an urban setting. The analyses reveal mixed results and do not suggest that the spatial regression approach unambiguously outperforms extrapolation projections.     

A method for deriving mortality estimates from incomplete vital Statistics

Summary Although they are available in many developing countries vital registration records are very little used for mortality estimation which is still mainly based on census returns. However, defective death records may yield accurate estimations of mortality. This procedure requires few data only; a sex-age distribution of the population (preferably at the middle of a period) and a sexage distribution of deaths, either derived from vital records or from census returns to questions relating to deaths during the preceding twelve months. This method is based on the observation that for a fixed age structure of the population, there is a one-one relation between the age structure of deaths (measured by the proportion of deaths at older ages) and the level of mortality (measured by the death rate above a certain minimum age). It is assumed that at ages above this minimum the rate of underregistration of deaths does not vary significantly with age. Therefore, the age distribution of registered deaths makes it possible to estimate the true proportion of deaths at older ages. This in its turn will permit the estimation of the true level of mortality, because of the relation which exists between age structure of deaths and level of mortality. The true level is then compared with the observed, to estimate the rate of underregistration, and observed age-specific death rates can be adjusted in the light of this knowledge.     

New regression formula to estimate the whole population for recapture-addicted small mammals

Summary We have devised a census formula of curvilinear regression suitable for capture-recapture data of recapture-addicted populations of the Japanese field vole (Microtus montebelli) obtained under a grid-plan with single-catch traps in order to estimate the whole population. The equation is founded on the assumption that the trappable population on the initial day is increased in way of an exponential curve until it reaches to the whole during one trapping period. The effect of trap-preoccupation by marked and multiple collisions is considered in the formula. As a result of its application to field data of the vole, it has turned out that the equation is required for the data gained under the trapping plan with trap spacing 10m, but not for those under the plan with spacing 5m, to estimate the whole. A convenient method of analysis of the formula is offered here, but we have been yet unable to introduce assymptotic variance of estimates. Contributions from JIBP-PT No. 19, carried out by the grant from the expenditure of Education Department to the specific study on “Dynamics of Biosphere”.