The main data provider is Statistics Portugal (INE), the central body responsible for the production and dissemination of official statistics in Portugal. Most data are freely available at INE’s website (Statistics Portugal, 2018) or can be requested free of charge to the same institution through an online form (Statistics Portugal, 2018). However, for specific datasets, such as International Trade and Industrial Production, microdata is only available to Approved Researchers.
For employment data, the main source is Gabinete de Estratégia e Planeamento (Ministério do Trabalho, Solidariedade e Segurança Social). The data request should be sent by email (Ministério do Trabalho, Solidariedade e Segurança Social), and fees may apply. Costs for employment data represented more than 90% of total data gathering costs for the Leiria UMAn model. These data are not subject to any confidential agreement.
In most cases, data is available at NUTS 2 level (corresponding to Região Centro) with the product disaggregation level requested in the Manual for data collection from D2.1 Model architecture report. The exception is most waste and emissions data, which could only be obtained at country level. For this, the Portuguese Environmental Agency – APA was contacted, but they were unable to provide more disaggregated data. At NUTS 3 level (corresponding to Região de Leiria) and LAU 1009 (Leiria municipality), data could only be collected for international transport, urban waste collection, mining and quarries production, population, and employment.
Overall, data could be obtained for at least a 10-year time series, except for most waste data, which are only available from 2008 onwards.
As regards classification systems, the following should be noted:
For some datasets (mainly primary production), it is not clear which classification system is used;
For mineral extraction and industrial production, data is provided in “National product list” (NPL) codes instead of Prodcom codes. Since there is no correspondence table between NLP and Prodcom, the automatic conversion is not easy. For mineral extraction, it was performed manually, by eliminating the 5th, 10th, 11th and 12th digits of the NLP codes. An automatic procedure should be set up to make the conversion for larger datasets (e.g. industrial production);
For economic activities, the Portuguese classification system (CAE Rev. 3) corresponds to NACE Rev. 2. The correspondence is direct, except that CAE has a 5-digit level, whilst NACE stops at 4 digits; and at 4-digit level, CAE has an extra class resulting from the disaggregation of “551 Hotels and similar accommodations” into two (with  and without  restaurant).
Credits: Câmara Municipal de Leiria
Power Point Presentation