Learn R Programming

SamplingStrata (version 1.5-5)

Optimal Stratification of Sampling Frames for Multipurpose Sampling Surveys

Description

Tools for the optimization of stratified sampling design. It determines a stratification of a sampling frame that minimizes sample cost while satisfying precision constraints in a multivariate and multidomain context. The approach relies on a genetic algorithm; each candidate partition of the frame is an individual whose fitness is evaluated via the Bethel-Chromy allocation to meet target precisions. Functions support analysis of optimization results, labeling of the frame with new strata, and drawing a sample according to the optimal allocation. Algorithmic components adapt code from the 'genalg' package. See M. Ballin and G. Barcaroli (2020) "R package SamplingStrata: new developments and extension to Spatial Sampling" .

Copy Link

Version

Install

install.packages('SamplingStrata')

Monthly Downloads

1,367

Version

1.5-5

License

GPL (>= 2)

Issues

Pull Requests

Stars

Forks

Repository

https://github.com/barcaroli/SamplingStrata/

Homepage

https://barcaroli.github.io/SamplingStrata/

Maintainer

Giulio Barcaroli

Last Published

October 8th, 2025

Functions in SamplingStrata (1.5-5)

Checks the inputs to the package: dataframes "errors", "strata" and "sampling frame"

Dataset 'nations'

Optimization of the stratification of a sampling frame given a sample survey

Function that allows to calculate a heteroscedasticity index, together with associate prediction variance, to be used by the optimization step to correctly evaluate the standard deviation in the strata due to prediction errors.

Precision constraints (maximum CVs) as input for Bethel allocation

Best stratification of a sampling frame for multipurpose surveys

buildStrataDFSpatial

Builds the "strata" dataframe containing information on target variables Y's distributions in the different strata, starting from sample data or from a frame

selectSampleSystematic

Selection of a stratified sample from the frame with systematic method

Selection of a stratified sample from the frame with srswor method

Plotting sampling rates in the different strata for each domain in the solution.

Dataframe containing information on strata in the frame

prepareSuggestion

Prepare suggestions for optimization with method = "continuous" or "spatial"

selectSampleSpatial

Selection of geo-referenced points from the frame

Procedure to apply Bethel algorithm and select a sample from given strata

Information on strata structure

Precision constraints (maximum CVs) as input for Bethel allocation

Dataframe containing information on strata in the swiss municipalities frame

Updates the initial frame on the basis of the optimized stratification

Execution and compared evaluation of optimization runs

Plot bivariate distibutions in strata

Assigns new labels to atomic strata on the basis of the optimized aggregated strata

optimizeStrata2

Best stratification of a sampling frame for multipurpose surveys (only with continuous stratification variables)

Allows to transform a continuous variable into a categorical ordinal one by applying a modified version of the k-means clustering function in the 'stats' package.

Dataframe containing information on all units in the population of reference that can be considered as the final sampling unit (this example is related to Swiss municipalities)

swissmunicipalities

The Swiss municipalities population

optimizeStrataSpatial

Best stratification of a sampling frame for multipurpose surveys considering also spatial correlation

aggrStrataSpatial

Builds the "strata" dataframe containing information on target variables Y's distributions in the different strata, starting from a frame where units are spatially correlated.

Builds the "strata" dataframe containing information on target variables Y's distributions in the different strata, starting from a frame

Multivariate optimal allocation

Initial solution obtained by applying kmeans clustering of atomic strata

Adjustment of the sample size in case it is externally given

buildFrameSpatial

Builds the "sampling frame" dataframe from a dataset containing information all the units in the population of reference including spatial

Builds the "sampling frame" dataframe from a dataset containing information on all the units in the population of reference

KmeansSolution2

Initial solution obtained by applying kmeans clustering of frame units

Expected coefficients of variation of target variables Y

Evaluation of the solution produced by the function 'optimizeStrata' by selecting a number of samples from the frame with the optimal stratification, and calculating average CV's on the target variables Y's.

assignStrataLabel

Function to assign the optimized strata labels

Builds the "strata" dataframe containing information on target variables Y's distributions in the different strata, starting from sample data or from a frame

KmeansSolutionSpatial

Initial solution obtained by applying kmeans clustering of frame units