D. SALGADO (david.salgado.fernandez@ine.es)
Dept. Methodology and Development of Statistical Production, Statistics Spain (INE)
M.E. ESTEBAN (elisa.esteban.segurado@ine.es)
Dept. Methodology and Development of Statistical Production, Statistics Spain (INE)
S. SALDAÑA (soledad.saldana.diaz@ine.es)
Dept. Methodology and Development of Statistical Production, Statistics Spain (INE)
Abstract
In the quest to industrialise the statistical production at Statistics Spain (INE) we have recently proposed to follow the principles of functional modularity to implement a modern production system. We conceive such a modern system as a set of layers ranging from the definition of the statistical needs or statistical problem to a final user interface, going over the statistical method(s) and the software implementation both (i) following both the functional and object-oriented paradigms and (ii) at a scripting level thereof. We have developed a collection of R packages providing a first proof of concept regarding an optimization-based approach to selective editing proposed and followed at Statistics Spain (INE) in several monthly Short-Term Business Statistics. All our R packages make an intensive use of the S4 object-oriented system pursuing a direct implementation of functional modularity principles with a strong methodological foundation of the design of modules. We present the current status of the project focusing on the implementation of the statistical methods going from small-scale functions in diverse R packages to survey-specific scripts in actual production.
Key words: R pack
JEL Classification: C10, C15, C18