Skip to content

eurostat/SAE_citystat

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

14 Commits
 
 
 
 
 
 
 
 

Repository files navigation

SAE_citystat

Experimental tools for Small Area Estimation (SAE) for city statistics and other functional areas.

The material provided herein will enable you to reproduce the experiments presented in Eurostat statistical working paper on Small area estimation for city statistics and other functional geographies (cite this source code or the reference's doi: 10.2785/898627). Further details see the in the working paper (below).

Description

The source code is provided as is in the r_code/ folder so as to repeat the simulation. The code installs the required packages if it is not available. It also prompts and downloads the AMELIA dataset version 0.2.3 which was used in the working paper.

It is important that the codes (numbered from 00_... through 05_... and performing individual steps of the simulation) including the contained subfolder r_code/Functions are kept together in one single folder.

Usage

The whole simulation starts from the first script (r_code/00_MAIN.R). It contains an interactive routing of the user through the whole procedure. Along the way, the user is asked whether she/he wishes to download the AMELIA dataset from the respective website. Also the code can be run as batch script in this case the data automatically downloaded if has not been previously and the default 20 iteraration is done in this case.

The codes contain numerous comments that should be really useful and the user will receive additional information in dialogue. Following all given the instructions, the complete simulation can be run.

ATTENTION: You have to take into account computation time, though. The simulation in the working paper was done on a large server infrastructure with approximately 1,200 cores. Reproducing the whole simulation study on just one computer will take many (!) days. Accordingly, it is recommended that the routine is just tested with a few simulation runs (default 20). This should suffice to get a grasp of the general procedure. Obviously, the results reached after only 20 simulation runs will not be the same as the final results given in the respective papers (which are based on many more simulation runs).

About

authors Münnich R., Burgard J. B., Ertz F., Lenau S., Manecke J., and Merkle H.
version 1.0
status 2019 – closed
license EUPL (cite the source code or the reference above!)

References

About

Simulation in R for Small Area Estimation (SAE) using the AMELIA dataset

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Languages