I've been working on this lately:

Stats analysis is an application which utilizes federated SPARQL queries to gather RDF Data Cube observation values from different endpoints based only on user's selections (independent, dependent variables, and reference period). The R language for statistical computing is employed to perform statistical analysis and visualizations. The Shiny application and server bridges the front-end Web user interface with R on the server-side in order to compare statistical observations, and stores analysis results in RDF for future research.

* http://stats.270a.info/
* http://csarven.ca/linked-statistical-data-analysis

It is version 0.1 (or maybe even less). So go easy.

Basically it gathers data from different statistical sources and does some analysis on it. Here is one that's most likely not interesting, but it will give you an idea:

http://stats.270a.info/analysis/worldbank:SP.DYN.IMRT.IN/transparency:CPI2009/year:2009

Independent variable from World Bank
Dependent variable from Transparency International
Reference period year 2009

Currently, it looks into following Linked Datasets:

* World Bank (only Indicators)
* European Central Bank
* OECD
* FAO of UN
* Transparency International

It can be extended simply by adding more query endpoints.

I plan to add IMF, BIS, and Eurostat soon.

Provenance is always kept intact (see "oh yeah?" link).

See also: http://270a.info/