Machine learning algorithms such as Random Forest (RF) are being increasingly applied on traditionally geographical topics such as population estimation. Even though RF is a well performing and generalizable algorithm, the vast majority of its implementations is still 'aspatial' and may not address spatial heterogenous processes. At the same time, remote sensing (RS) data which are commonly used to model population can be highly spatially heterogeneous. From this scope, we present a novel geographical implementation of RF, named Geographical Random Forest (GRF) as both a predictive and exploratory tool to model population as a function of RS covariates. GRF is a disaggregation of RF into geographical space in the form of local sub-models. From the first empirical results, we conclude that GRF can be more predictive when an appropriate spatial scale is selected to model the data, with reduced residual autocorrelation and lower Root Mean Squared Error (RMSE) and Mean Absolute Error (MAE) values. Finally, and of equal importance, GRF can be used as an effective exploratory tool to visualize the relationship between dependent and independent variables, highlighting interesting local variations and allowing for a better understanding of the processes that may be causing the observed spatial heterogeneity.
Frequency of contacts with the family is an indicator of the strength of intergenerational exchange and potential support for older people. Although the availability of children clearly represents a constraint on potential family support, the extent of interaction with and support received from children depends on factors other than demographic availability alone. This study examined the effects of socio-economic and demographic variables on weekly contacts with children in Great Britain, Italy, Finland and The Netherlands using representative survey data which included information on availability of children and extent of contact. Our results confirm the higher level of parent adult-child contact in Italy than in northern European countries, but levels of contact in all the countries considered were high. Multivariate analysis showed that in most countries characteristics such as divorce were associated with a reduced probability of contact between fathers and children; in Finland this also influenced contact between mothers and children. Analyses are also included of possible future scenarios of contact with children that combine the observed effects of the explanatory variables with hypothetical changes in population distribution.
We investigate mortality differentials by marital status among older age groups using a database of mortality rates by marital status at ages 40 and over for seven European countries with 1 billion person-years of exposure. The mortality advantage of married people, both men and women, continues to increase up to at least the age group 85-89, the oldest group we are able to consider. We find the largest absolute differences in mortality levels between marital status groups are at high ages, and that absolute differentials are: (i) greater for men than for women; (ii) similar in magnitude across countries; (iii) increase steadily with age; and (iv) are greatest at older age. We also find that the advantage enjoyed by married people increased over the 1990s in almost all cases. We note that results for groups such as older divorced women need to be interpreted with caution.
The aim of the research that prompted this series of papers was to provide the Office of the Deputy Prime Minister (ODPM) with a migration modelling system, MIGMOD, potentially capable of informing the national government about policies that might have an impact on population redistribution within England and Wales. Some current concerns, for example, include: the drift of population to the South East, with concomitant increases in land pressure and house prices in that region; long-standing regional problems that are compounded by the loss of skilled population from economically depressed areas; and the trend of counterurbanisation involving a net loss of internal migrants from major urban areas to surrounding rural areas.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.