Publications

Income and Expenditure Accounts Technical Series

Constructing Provincial Time Series: A Discussion of Data Sources and Methods

2. Principles

Warning View the most recent version.

Archived Content

Information identified as archived is provided for reference, research or recordkeeping purposes. It is not subject to the Government of Canada Web Standards and has not been altered or updated since it was archived. Please "contact us" to request a format other than those available.

The new provincial dataset is designed to meet four criteria: quality, relevance, expandability, and ability to be updated. The four criteria are discussed in order of importance.

The first criteria is quality. To meet this goal, official sources are used whenever possible, notably, estimates from Statistics Canada, from the Dominion Bureau of Statistics, and from the Historical Statistics of Canada. If official estimates are not available, sources such as academic studies are used to improve coverage and time span. In all cases, the data source and how it was used are documented to allow for replication and verification.

The second criteria is relevance. To accomplish this, modern estimates from the Canadian System of National Accounts are used, and the data are linked in such a way that they can be updated as additional data points become available. Consequently, all variables are produced so as to link historical data to modern data. Specifically, historical estimates are adjusted to produce back-cast estimates that do not exhibit structural breaks or level shifts when changes are made in the vintage of data employed.

Third, the dataset must be expandable―over time and across the indicators. All expansions that result in new data points going forward through time need to be made in a timely fashion. All expansions that lead to new variables being added to the dataset, or that increase the span of the data historically need to be published with documentation.

Fourth, the dataset should be open to revision. Because it is a research dataset, knowledge acquired through its use can be applied to improve quality. For instance, better data sources may be located, and if the dataset is expanded, the need for consistent estimates across a range of variables may necessitate new and different linking procedures. All revisions should be released with documentation.

Date modified: