Estimating Small-Area Populations by Age and Sex Using Spatial Interpolation and Statistical Inference Methods

Cai, Qiang; Rushton, Gérard; Bhaduri, Budhendra; Bright, Edward A.; Coleman, P.R.

doi:10.1111/j.1467-9671.2006.01013.x

Cited by 39 publications

(26 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Three indices from the regression analysis are used for the comparison, i.e. regression coefficient (b), coefficient of determination (R 2 ) and mean square error (MSE) (Cai et al 2006). The regression coefficients are all smaller than that with the largest value from stepwise regression algorithm.…”

Section: Resultsmentioning

confidence: 99%

Integration of GP and GA for mapping population distribution

Liao

Wang

Meng

et al. 2010

International Journal of Geographical Information Science

View full text Add to dashboard Cite

Mapping population distribution is an important field of geographical and related research because of the frequent need to combine spatial data representing socio-demographic information across various incompatible spatial units. However, the research may become very complex and difficult when a population in multiple places is estimated by various factors. Previous efforts in the field have contributed to the selection of appropriate independent variables and the creation of different population models. However, the level of accuracy obtainable with these studies is limited by the spatial heterogeneity of population distribution within the individual census districts, particularly in large rural areas. A high-accuracy modelling method for population estimation based on integration of Genetic Programming (GP) and Genetic Algorithms (GA) with Geographic Information Systems (GIS) is presented in this paper. GIS was applied to identify and quantify a set of natural and socioeconomic factors which contributed to population distribution, and then GP and GA were used to build and optimise the population model to automatically transform census population data to regular grids. The study indicated that the proposed method performed much better than the stepwise regression analysis and adapted gravity model methods in estimating the population of both urban and rural areas. More importantly, this proposed method could provide a single, unified approach to mapping population distribution in various areas because the paradigms of these algorithms are general.

show abstract

Section: Resultsmentioning

confidence: 99%

Integration of GP and GA for mapping population distribution

Liao

Wang

Meng

et al. 2010

International Journal of Geographical Information Science

View full text Add to dashboard Cite

show abstract

“…Numerous studies consider topographic features as the basic factors that influence the population distribution such as (1) altitude (Mennis 2003, Yue et al 2003, Flowerdew et al 2007, Langford et al 2008); (2) slope (Cai et al 2006, Liao et al 2010; and (3) hydrology (Yuan et al 1997, Flowerdew et al 2007. Transport network, school catchment, markets, service centres and land-use classification are all important factors that are also taken frequently as major input variables (for instance, Harvey 2002, Mennis 2003, Balk et al 2006, Langford et al 2008, Liu et al 2008.…”

Section: Input Variables Chosen From the Planning Perspectivementioning

confidence: 99%

The viability of the genetic algorithm geodemographic modelling in the case of Korean DIF zoning

Choei

2012

International Journal of Geographical Information Science

View full text Add to dashboard Cite

Development impact fees (DIF) are used for the provision of public infrastructure services to adequately serve new developments. Korean Ministry of Land, Transport, and Maritime Affairs introduced the DIF zoning in 2008 and, like its US counterpart, it requires Korean localities to designate specific districts called 'DIF zones' based on the local population growth rate. This study examines the genetic algorithms as a method for DIF zoning-related geodemographic modelling using the Korean National Geographic Information Systems as property-level ancillary data. A borough of Hwasung City is taken as the case area since the city internally collected population data by ward-level enumeration areas for DIF zoning in 2008. The gridded population map is built from this source enumeration area population data to select the training dataset. The functional form of the genetic algorithm model has been formulated to have a hierarchical weighting system in which categorical weights of variable groups and individual weights of subordinate variables are sought bilaterally. The model is run by a carefully pretested set of reproductive plan parameters and the consequences are compared with the conventional regression models. It is found that the genetic algorithm solutions are quite comparable to the ones obtained by the regression methods, and it seemed, in this regard, worth adopting the two approaches simultaneously in a complementary manner to take unique advantages of each other either in facilitating the analysis processes or in obtaining more promising outputs for the DIF zoning as well as other geodemographic applications.

show abstract

“…Exposure analysis for public health and socio-environmental (environmental justice) studies can have tremendous benefits from LandScan USA data. In spatial epidemiology and disease (cancer) mapping, the utility of LandScan USA has been well illustrated (Cai et al 2006). It is realized that activity based and time specific high resolution population distribution data will be of great advantage for socioeconomic (research and commercial) applications for evaluating the potential for Location Based Services (LBS) such as access to healthcare or coverage for wireless and cellular phones.…”

Section: Cross Validation Using Imagerymentioning

confidence: 99%

“…validate the number of people predicted for each cell), it is possible to assess the locational accuracy and precision of the model and data. With very high resolution orthophotographs, 964 geocoded house locations were compared with an earlier version of LandScan USA data in Iowa (Cai et al 2006). Using a spatial sensitivity filter of 90 m, the analysis indicated 72.5% accuracy in predicting populated cells over house locations and 99% accuracy in predicting unpopulated areas.…”

Section: Cross Validation Using Imagerymentioning

confidence: 99%

LandScan USA: a high-resolution geospatial and temporal modeling approach for population distribution and dynamics

et al. 2007

Self Cite

View full text Add to dashboard Cite

High-resolution population distribution data are critical for successfully addressing important issues ranging from socio-environmental research to public health to homeland security, since scientific analyses, operational activities, and policy decisions are significantly influenced by the number of impacted people. Dasymetric modeling has been a well-recognized approach for spatial decomposition of census data to increase the spatial resolution of population distribution. However, enhancing the temporal resolution of population distribution poses a greater challenge. In this paper, we discuss the development of LandScan USA, a multi-dimensional dasymetric modeling approach, which has allowed the creation of a very high-resolution population distribution data both over space and time. At a spatial resolution of 3 arc seconds (*90 m), the initial LandScan USA database contains both a nighttime residential as well as a baseline daytime population distribution that incorporates movement of workers and students. Challenging research issues of disparate and misaligned spatial data and modeling to develop a database at a national scale, as well as model verification and validation approaches are illustrated and discussed. Initial analyses indicate a high degree of locational accuracy for LandScan USA distribution model and data. High-resolution population data such as LandScan USA, which describes both distribution and dynamics of human population, clearly has the potential to profoundly impact multiple domain applications of national and global priority.

show abstract

Estimating Small-Area Populations by Age and Sex Using Spatial Interpolation and Statistical Inference Methods

Cited by 39 publications

References 22 publications

Integration of GP and GA for mapping population distribution

Integration of GP and GA for mapping population distribution

The viability of the genetic algorithm geodemographic modelling in the case of Korean DIF zoning

LandScan USA: a high-resolution geospatial and temporal modeling approach for population distribution and dynamics

Contact Info

Product

Resources

About