SAE Unit/area Models and Methods for Estimation in R
Version 1.3.0 will add population and prevalence simulation functionality based on population frame and population density information, along with methods for aggregating pixel level prevalences and populations to the areal level. These functions are currently available in the dev branch.
Updates will include: + New major functions. * simSPDE
: Simulates from the SPDE spatial model. * simPopSPDE
: Simulates population in a country based on SPDE spatial model and population frame information. * simPopPixel
: Simulates population in a country based on custom spatial model and population frame information. + New minor functions. * aggPixelPreds
and areaPopToArea
: Functions for aggregating populations and prevalences to the areal level. * Functions for setting pixellated grid and population density grid. * Functions for adjusting population density grid based on population frame information. * Functions for projections to use with Kenya maps and for getting Admin areas given spatial locations. + New internal functions for helping with population simulation. + New datasets * kenyaMaps
: Kenya administrative area map shapefiles, and a triangular mesh for the SPDE model for Kenya. * kenyaPopulationData
: General and neonatal population frames for Kenya, and general and neonatal population density information for Kenya. * popMatKenya
and popMatKenyaNeonatal
: General and neonatal population density grids at 5km resolution.
smoothSurvey
to implement popular SAE methods. Syntax change to the function.smoothDirect
and smoothCluster
to fit space-only models.smoothSurvey
, smoothDirect
, and smoothCluster
now returns S3 classed objects.Version 1.0.0 contains many major updates from the previous versions: + Major updates to functions. + fitGeneric
is now smoothSurvey
+ fitINLA
is now smoothDirect
+ fitINLA2
is now smoothCluster
+ More extensions in both smoothed direct and cluster level models. + Major changes to how temporal models are specified with time.model
and st.time.model
. + More interpretable parameterization of slope and random slopes. + More visualization options. + Note: Previous function name and argument syntax remain to work as before, but may not receive high priority in maintenance in the future.
+ Many minor improvements in functions. + Better model summary message. + Removed unnecessary function arguments, e.g., geo
in various functions. + Removed the requirement to repeated specifying Amat
, year_label
and year_range
. Now they are only required in the model fitting stage. + New vignettes.
Version 0.3.0 contains some major updates from the previous versions. Some of the substantial changes to existing functions are listed here. For a complete log of changes, see the News section.
countrySummary
is now getDirect
cuontrySummary_mult
is now getDirectList
fitspace
is now fitGeneric
projINLA
is now getSmooth
getDiag
: produce diagnostic plots for the fitted model.getAdjusted
: produce adjusted estimates for a fitted model.getAmat
: automatic extract spatial adjacency matrix from the polygon file.hatchPlot
: plot variables on a map with hatching indicating the width of the credible interval. +New methodsfitINLA2
: implements new smoothing methods based on binomial models at cluster level.To cite the SUMMER package in publications use
@Manual{li2020space,
title = {Space-Time Smoothing of Demographic and Health Indicators using the R Package SUMMER},
author = {Zehang R Li and Bryan D Martin and Tracy Q Dong and Geir-Arne Fuglstad and Jessica Godwin and John Paige and Andrea Riebler and Samuel Clark and Jon Wakefield},
year = {2020},
journal = {arXiv preprint}
}
To cite specific version of the SUMMER package use + v1.0.0
@Manual{summer2020,
title = {SUMMER: Spatio-Temporal Under-Five Mortality Methods for Estimation},
author = {Zehang R Li and Bryan D Martin and Yuan Hsiao and Jessica Godwin and Jon Wakefield and Samuel J Clark and Geir-Arne Fuglstad and Andrea Riebler},
year = {2020},
note = {R package version 1.0.0},
}
@Manual{summer2019,
title = {SUMMER: Spatio-Temporal Under-Five Mortality Methods for Estimation},
author = {Bryan D Martin and Zehang R Li and Yuan Hsiao and Jessica Godwin and Jon Wakefield and Samuel J Clark and Geir-Arne Fuglstad and Andrea Riebler},
year = {2019},
note = {R package version 0.3.0},
}
The package is now available on CRAN. The easiest way to download is to install directly using the code below.
To download the development version of the SUMMER package, use the code below.
Examples of most of the main functions are described in the several vignettes listed on https://cran.r-project.org/package=SUMMER.
If you encounter a bug or would like make a change request, please file it as an issue here.