Tempus3 database

Contains information on all the INE's short-term statistical operations, those for which the results dissemination frequency is less than one year, and certain structural statistical operations.

The list of operations in Tempus3 changes as they are integrated into the database. The operations available in the Tempus3 system can be viewed through the following URL: https://servicios.ine.es/wstempus/js/EN/OPERACIONES_DISPONIBLES.

Tempus3 is a relational database that uses a set of objects organised in a hierarchy to store and manage statistical information. Its main element is a time series, the only object to which data are associated. The rest of the elements emerge around it. Operations and statistical tables will be container objects of time series.

Conceptos fundamentales

To use the service correctly, it is necessary to describe the following concepts, whose identifiers will be the inputs (input values) in the construction of URLs.

Variables

A variable is a characteristic that can fluctuate and whose variation is likely to take on different values.

The variables or lists of values contained in Tempus3 and used in dissemination are common to all statistical operations, i.e. they are not duplicated in the system, their Tempus3 identifier (Id) and their descriptors are unique. The variables can be looked up with the function VARIABLES.

Examples of variables are lists: COICOP Groups, Sex, Autonomous Communities, Provinces,...

Values

Values are the states that a given variable can have. For example, the variable Provinces contains the values: Áraba/Álava, Albacete, Alicante/Alacant,...

The values contained in Tempus3 and used in the dissemination are common to all statistical operations, i.e. they are not duplicated in the system, their Tempus3 identifier (Id) and descriptors are unique. The values can be looked up with the function VARIABLE_VALUES.

Series

What are time series?

A set of observations measured at specific points in time, arranged chronologically and usually evenly spaced.

As the main entity of the Tempus3 database, the series has defining properties that do not change over time:

Unique identifier and series characteristics: id, name, periodicity, scale, unit, classification, decimal places,...

Tables

One of the advantages of this system is the ease of managing the different forms of presentation, which match the different ways in which the series can be grouped. One of them is grouping them in a table (or cube). It is the most one most often used by users, who access statistical tables through INEbase.

A table is the result of crossing groups of values contained in one or more variables, i.e. they are a grouping of time series defined by these groups.

For example, you navigate through INEbase to get to the table Indices by Autonomous Communities: general and ECOICOP group indices of the Consumer Price Index, Base 2021. Once the table has been accessed, it can be seen that it contains all the series resulting from the combination of the groups of values contained in the following variables:

{Autonomous Communities and Cities} x {ECOICOP groups} x {Data type}

As an entity, the table has its own characteristics such as its name, the periods it comprises and its unique identifier. You can obtain a list of all the tables of an operation with the function TABLES OPERATION.

Publication date

Data in Tempus3 are published at the series level and are associated with a moment in time or reference date (period/year)..

But when are the data published? When are they updated?

Each operation has one or several publications associated with it according to its different periodicities, e.g. there are two publications for the CPI, one with a monthly periodicity and one with an annual periodicity. These follow the INE publication calendar.

In this way, a publication contains the times at which the data of a statistical operation are published: publication dates.

With the publication identifier you can look up the publication date ofall statistical operations available in the publication calendar with the function PUBLICATION_DATE_PUBLICATION. Each of these publication dates has its own characteristics: