No abstract

No abstract

This article describes a project conducted in conjunction with the Central Statistics Office of Ireland in response to a planned national rollout of smart electricity metering in Ireland. We investigate how this new data source might be used for the purpose of official statistics production. This study specifically looks at the question of determining household composition from electricity smart meter data using both Neural Networks (a supervised machine learning approach) and Elastic Net Logistic regression. An overview of both classification techniques is given. Results for both approaches are presented with analysis. We find that the smart meter data alone is limited in its capability to distinguish between household categories but that it does provide some useful insights.

This paper looks at the evolving National Data Infrastructure in Ireland and describes developments to date before considering the emerging Census opportunity. The National Data Infrastructure concept is influenced by the ideas of Svein Nordbotten first introduced in the 1960s and by the representation of three pillars or linked registers (persons, business and property) presented by Lars Thygesen in the 1980s. The National Data Infrastructure is, in turn, laying a foundation for the modernisation of the Irish Statistical System. The statistical business register at the Central Statistics Office (CSO) led to the publication of a comprehensive system of business demography statistics for the first time in 2010. This statistical business register, the first pillar to be developed, is typical of many registers in the European Statistical System, in that it is based on the profiling of businesses from administrative data sources (typically sourced from tax authorities). Subsequently, the employer-employee tax returns from the tax authorities were used to provide insights into dynamics in the Irish Labour market. This allowed for the investigation of the flow of workers and jobs between firms and sectors. This underlying development is also of strategic value to the Irish Statistical System in that it facilitates linking the social statistics framework with the business statistics framework at a micro or person business level. CSO is currently investigating a system of population estimates based solely on administrative data sources. This project leverages on the recently developed person activity register-a register that captures a high level annual summary of each person's engagement on public administration systems. The person activity register is considered as the second base register required for the modernisation of the Irish Statistical System. Currently, a critical gap in the National Data Infrastructure is the absence of a standardised addressing system or postal code system. In 2015, Ireland will rollout a new postcode system and, unlike other jurisdictions, the postcode in Ireland will identify a letter box rather than a group of houses. The implementation and use of this postcode (for both business and persons) will facilitate the compilation of small area statistics (including population estimates). It could also contribute to the development of household composition based statistics through the linking of individuals via postcode. In effect, postcodes will provide the last of the three registers identified in the National Data Infrastructure. To date, Ireland has deployed a traditional census model whereby forms are dropped off and picked up from every household in the state. The implementation of letterbox based postcodes creates an opportunity for a new Census model. The options for a new model include any combination of internet response, postal response, administrative data or field/telephone follow up underpinned by a high quality address register developed from administrative data sources. The lowe...

No abstract