One important part of making official statistics is how to present the statistical information. The aim of data visualisation is to makes patterns in the data – such as changes, trends, deviations, correlations and outliers – clearer for users. It visualises the story in the figures. This report describes the potential pitfalls of using graphics and maps to present statistical data. It gives a number of guidelines for the correct presentation of statistical information, and refers to the available literature on this subject.
Statistics Netherlands is required by law to protect the privacy of its respondents as well as it can. This process, called statistical disclosure control, can be carried out in various ways. This report describes methods available at Statistics Netherlands for statistical disclosure control of both microdata and quantitative tables. In addtion, the protection of frequency tables and analysis results is discussed.
Benchmarking is a specific type of reconciliation problem in macro-integration. It consists of reconciling systems of accounts based on different periods (e.g. quarterly versus annual accounts). This report contains an extension of the multivariate Denton method for benchmarking (see also the report ‘Macro-integration: Data reconciliation’). The extension comprises soft restrictions, inequality restrictions and ratio restrictions.
Life tables describe the mortality and survival patterns of a population. On the basis of (annual) mortality rates for each age or age group, they provide information on parameters such as the number of survivors, the number of deaths and life expectancy.Statistics Netherlands publishes two types of life tables in StatLine: tables by average age on 1 January and tables by age on last birthday. This report examines the calculation methods used in the two types of tables.
Many time series at Statistics Netherlands suffer from seasonal or working-day effects. Energy use by households, for example, will be higher in the winter than in the summer. In the interpretation of month-on-month or quarter-on-quarter figures, the changes without these effects are important. Therefore Statistics Netherlands publishes both the series corrected for seasonal and working-day effects and the uncorrected series. The standard method used at Statistics Netherlands for seasonal and working-day correction is X12-ARIMA. This report discusses both the method and the software. It also contains an extensive example of the application of the X12-ARIMA program to a time series at Statistics Netherlands.
An index or index number combines relative changes in various variables into one figure. Price and volume indices are important indicators in economic statistics. They give aggregated information on changes in prices and volumes of goods and services. A well-known example is the consumer price index. This report examines the most often used index formulas and their characteristics.
Macro-economic statistics published by Statistics Netherlands must be consistent. The figure for GDP calculated according to the income approach, for example, must be the same as the figure calculated according to the production approach. To make macro-economic figures consistent, first of all large measurement errors are corrected, and then the remaining discrepancies are solved. The latter process is also called reconciliation. This report describes a number of mathematical reconciliation methods.