metadata | Economy Data Observatory

How We Add Value to Public Data With Imputation and Forecasting

Mon, 08 Nov 2021 10:00:00 +0100

Public data sources are often plagued by missng values. Naively you may think that you can ignore them, but think twice: in most cases, missing data in a table is not missing information, but rather malformatted information. This approach of ignoring or dropping missing values will not be feasible or robust when you want to make a beautiful visualization, or use data in a business forecasting model, a machine learning (AI) applicaton, or a more complex scientific model. All of the above require complete datasets, and naively discarding missing data points amounts to an excessive waste of information. In this example we are continuing the example a not-so-easy to find public dataset.

In the previous blogpost we explained how we added value by documenting data following the FAIR principle and with the professional curatorial work of placing the data in context, and linking it to other information sources, such as other datasets, books, and publications, regardless of their natural language (i.e., whether these sources are described in English, German, Portugese or Croatian). Photo: Jack Sloop.

Completing missing datapoints requires statistical production information (why might the data be missing?) and data science knowhow (how to impute the missing value.) If you do not have a good statistician or data scientist in your team, you will need high-quality, complete datasets. This is what our automated data observatories provide.

Why is data missing?

International organizations offer many statistical products, but usually they are on an ‘as-is’ basis. For example, Eurostat is the world’s premiere statistical agency, but it has no right to overrule whatever data the member states of the European Union, and some other cooperating European countries give to them. And they cannot force these countries to hand over data if they fail to do so. As a result, there will be many data points that are missing, and often data points that have wrong (obsolete) descriptions or geographical dimensions. We will show the geographical aspect of the problem in a separate blogpost; for now, we only focus on missing data.

Some countries have only recently started providing data to the Eurostat umbrella organization, and it is likely that you will find few datapoints for North Macedonia or Bosnia-Herzegovina. Other countries provide data with some delay, and the last one or two years are missing. And there are gaps in some countries’ data, too.

See the authoritative copy of the dataset.

This is a headache if you want to use the data in some machine learning application or in a multiple or panel regression model. You can, of course, discard countries or years where you do not have full data coverage, but this approach usually wastes too much information–if you work with 12 years, and only one data point is available, you would be discarding an entire country’s 11-years’ worth of data. Another option is to estimate the values, or otherwise impute the missing data, when this is possible with reasonable precision. This is where things get tricky, and you will likely need a statistician or a data scientist onboard.

What can we improve?

Consider that the data is only missing from one year for a particular country, 2015. The naive solution would be to omit 2015 or the country at hand from the dataset. This is pretty destructive, because we know a lot about the radio market turnover in this country and in this year! But leaving 2015 blank will not look good on a chart, and will make your machine learning application or your regression model stop.

A statistician or a radio market expert will tell you that you know more-or-less the missing information: the total turnover was certainly not zero in that year. With some statistical or radio domain-specific knowledge you will use the 2014, or 2016 value, or a combination of the two and keep the country and year in the dataset.

Our improved dataset added backcasted (using the best time series model fitting the country’s actually present data), forecasted (again, using the best time series model), and approximated data (using linear approximation.) In a few cases, we add the last or next known value. To give a few quantiative indicators about our work:

Increased number of observations: 65%
Reduced missing values: -48.1%
Increased non-missing subset for regression or AI: +66.67%

If your organization is working with panel (longitudional multiple) regressions or various machine learning applications, then your team knows that not havint the +66.67% gain would be a deal-breaker in the choice of models and punctuality of estimates or KPIs or other quantiative products. And that they would spent about 90% of their data resources on achieving this +66.67% gain in usability.

If you happen to work in an NGO, a business unit or a research institute that does not employ data scientists, then it is likely that you can never achieve this improvement, and you have to give up on a number of quantitative tools or visualizations. If you have a data scientist onboard, that professional can use our work as a starting point.

Can you trust our data?

We believe that you can trust our data better than the original public source. We use statistical expertise to find out why data may be missing. Often, it is present in a wrong location (for example, the name of a region changed.)

If you are reluctant to use estimates, think about discarding known actual data from your forecast or visualization, because one data point is missing. How do you provide more accurate information? By hiding known actual data, because one point is missing, or by using all known data and an estimate?

Our codebooks and our API uses the Statistical Data and Metadata eXchange documentation standards to clearly indicate which data is observed, which is missing, which is estimated, and of course, also how it is estimated. This example highlights another important aspect of data trustworthiness. If you have a better idea, you can replace them with a better estimate.

Our indicators come with standardized codebooks that do not only contain the descriptive metadata, but administrative metadata about the history of the indicator values. You will find very important information about the statistical method we used the fill in the data gaps, and even link the reliable, the peer-reviewed scientific, statistical software that made the calculations. For data scientists, we record the plenty of information about the computing environment, too-–this can come handy if your estimates need external authentication, or you suspect a bug.

Avoid the data Sisyphus

If you work in an academic institution, in an NGO or a consultancy, you can never be sure who downloaded the Annual detailed enterprise statistics for services (NACE Rev. 2 H-N and S95) Eurostat folder from Eurostat. Did they modify the dataset? Did they already make corrections with the missing data? What method did they use? To prevent many potential problems, you will likely download it again, and again, and again…

See our The Data Sisyphus blogpost.

We have a better solution. You can always rely on our API to import directly the latest, best data, but if you want to be sure, you can use our regular backups on Zenodo. Zenodo is an open science repository managed by CERN and supported by the European Union. On Zenodo, you can find an authoritative copy of our indicator (and its previous versions) with a digital object identifier, in this case, 10.5281/zenodo.5652118. These datasets will be preserved for decades, and nobody can manipulate them. You cannot accidentally overwrite them, and we have no backdoor access to modify them.

Are you a data user? Give us some feedback! Shall we do some further automatic data enhancements with our datasets? Document with different metadata? Link more information for business, policy, or academic use? Please give us any feedback!

How We Add Value to Public Data With Better Curation And Documentation?

Mon, 08 Nov 2021 09:00:00 +0100

In this example, we show a simple indicator: the Turnover in Radio Broadcasting Enterprises in many European countries. This is an important demand driver in the Music economy pillar of our Digital Music Observatory, and important indicator in our more general Cultural & Creative Sectors and Industries Observatory. Of course, if you work with competition policy or antitrust, than any industry may be interesting to you–but not all of them are well-serverd with data.

This dataset comes from a public datasource, the data warehouse of the European statistical agency, Eurostat. Yet it is not trivial to use: unless you are familiar with national accounts, you will not find this dataset on the Eurostat website.

The data can be retrieved from the Annual detailed enterprise statistics for services NACE Rev.2 H-N and S95 Eurostat folder.

Our version of this statistical indicator is documented following the FAIR principles: our data assets are findable, accessible, interoperable, and reusable. While the Eurostat data warehouse partly fulfills these important data quality expectations, we can improve them significantly. And we can also improve the dataset, too, as we will show in the next blogpost.

Findable Data

Our data observatories add value by curating the data–we bring this indicator to light with a more descriptive name, and we place it in a domain-specific context with our Digital Music Observatory and Cultural & Creative Sectors and Industries Observatory and a policy-specific context with our Competition Data Observatory and Green Deal Data Observatory. While many people may need this dataset in the creative sectors, or among cultural policy designers, most of them have no training in working with national accounts, which imply decyphering national account data codes in records that measure economic activity at a national level. Our curated data observatories bring together many available data around important domains. Our Digital Music Observatory, for example, aims to form an ecosystem of music data users and producers.

We added descriptive metadata that help you find our data and match it with other relevant data sources.

We added descriptive metadata that help you find our data and match it with other relevant data sources. For example, we add keywords and standardized metadata identifiers from the Library of Congress Linked Data Services, probably the world’s largest standardized knowledge library description. This ensures that you can find relevant data around the same key term (radio broadcasting) in addition to our turnover data. This allows connecting our dataset unambiguosly with other information sources that use the same concept, but may be listed under different keywords, such as Radio–Broadcasting, or Radio industry and trade, or maybe Hörfunkveranstalter in German, or Emitiranje radijskog programa in Croatian or Actividades de radiodifusão in Portugese.

Accessible Data

Our data is accessible in two forms: in csv tabular format (which can be read with Excel, OpenOffice, Numbers, SPSS and many similar spreadsheet or statistical applications) and in JSON for automated importing into your databases. We can also provide our users with SQLite databases, which are fully functional, single user relational databases.

Tidy datasets are easy to manipulate, model and visualize, and have a specific structure: each variable is a column, each observation is a row, and each type of observational unit is a table. This makes the data easier to clean, and far more easier to use in a much wider range of applications than the original data we used. In theory, this is a simple objective, yet we find that even governmental statistical agencies–and even scientific publications–often publish untidy data. This poses a significant problem that implies productivity loses: tidying data will require long hours of investment, and if a reproducible workflow is not used, data integrity can also be compromised: chances are that the process of tidying will overwrite, delete, or omit a data or a label.

Tidy datasets are easy to manipulate, model and visualize, and have a specific structure: each variable is a column, each observation is a row, and each type of observational unit is a table.

While the original data source, the Eurostat data warehouse is accessible, too, we added value with bringing the data into a tidy format. Tidy data can immediately be imported into a statistical application like SPSS or STATA, or into your own database. It is immediately available for plotting in Excel, OpenOffice or Numbers.

Interoperability

Our data can be easily imported with, or joined with data from other internal or external sources.

All our indicators come with standardized descriptive metadata, and statistical (processing) metadata. See our API

All our indicators come with standardized descriptive metadata, following two important standards, the Dublin Core and DataCite–implementing not only the mandatory, but the recommended descriptions, too. This will make it far easier to connect the data with other data sources, e.g. turnover with the number of radio broadcasting enterprises or radio stations within specific territories.

Our passion for documentation standards and best practices goes much further: our data uses Statistical Data and Metadata eXchange standardized codebooks, unit descriptions and other statistical and administrative metadata.

Reuse

All our datasets come with standardized information about reusabililty. We add citation, attribution data, and licensing terms. Most of our datasets can be used without commercial restriction after acknowledging the source, but we sometimes work with less permissible data licenses.

In the case presented here, we added further value to encourage re-use. In addition to tidying, we significantly increased the usability of public data by handling missing cases. This is the subject of our next blogpost.

Metadata

Wed, 07 Jul 2021 00:00:00 +0000

Adding metadata exponentially increases the value of data. Did your region add a new town to its boundaries? How do you adjust old data to conform to constantly changing geographic boundaries? What are some practical ways of combining satellite sensory data with my organization’s records? And do I have the right to do so? Metadata logs the history of data, providing instructions on how to reuse it, also setting the terms of use. We automate this labor-intensive process applying the FAIR data concept.

In our observatory we apply the concept of FAIR (findable, accessibe, interoperable, and reusable digital assets) in our APIs and in our open-source statistical software packages.

The hidden cost item

Metadata gets less attention than data, because it is never acquired separately, it is not on the invoice, and therefore it remains an a hidden cost, and it is more important from a budgeting and a usability point of view than the data itself. Metadata is responsible for industry non-billable hours or uncredited working hours in academia. Poor data documentation, lack of reproducible processing and testing logs, inconsistent use of currencies, keywords, and storing messy data make reusability and interoperability, integration with other information impossible.

FAIR Data and the Added Value of Rich Metadata we introduce how we apply the concept of FAIR (findable, accessibe, interoperable, and reusable digital assets) in our APIs.

Organizations pay many times for the same, repeated work, because these boring tasks, which often comprise of tens of thousands of microtasks, are neglected. Our solution creates automatic documentation and metadata for your own historical internal data or for acquisitions from data vendors. We apply the more general Dublin Core and the more specific, mandatory and recommended values of DataCite for datasets – these are new requirements in EU-funded research from 2021. But they are just the minimal steps, and there is a lot more to do to create a diamond ring from an uncut gem.

Map your data: bibliographis, catalogues, codebooks, versioning

Updating descriptive metadata, such as bibliographic citation files, descriptions and sources to data files downloaded from the internet, versioning spreadsheet documents and presentations is usually a hated and often neglected task withing organization, and rightly so: these boring and error-prone tasks are best left to computers.

Already adjusted spreadsheets are re-adjusted and re-checked. Hours are spent on looking for the right document with the rigth version. Duplicates multiply. Already downloaded data is downloaded again, and miscategorized, again. Finding the data without map is a treasure hunt. Photo: © N.

The lack of time and resources spend on documentation over time reduces reusability and significantly increases data processing and supervision or auditing costs.

Our observatory metadata is compliant with the Dublin Core Cross-Domain Attribute Set metadata standard, but we use different formatting. We offer simple re-formatting from the richer DataCite to Dublin Core for interoperability with a wider set of data sources.
We use all mandatory DataCite metadata fields, all the the recommended and optional ones.
It complies with the tidy data principles.

In other words: very easy to import into your databases, or join with other databases, and the information is easy to find. Corrections, updates can automatically managed.

What happened with the data before?

We are creating Codebooks that are following the SDMX statistical metadata codelists, and resemble the SMDX concepts used by international statistical agencies. (See more technical information here.)

Small organizations often cannot afford to have data engineers and data scientists on staff, and they employ analysts who work with Excel, OpenOffice, PowerBI, SPSS or Stata. The problem with these applications is that they often require the user to manually adjust the data, with keyboard entries or mouse clicks. Furthermore, they do not provide a precise logging of the data processing, manipulation history. The manual data processing and manipulation is very error prone and makes the use of complex and high value resources, such as harmonized surveys or symmetric input-output tables, to name two important source we deal with, impossible to use. The use of these high-value data sources often requires tens of thousands of data processing steps: no human can do it faultlessly.

What is even more problematic that simple applications for analysis do not provide a log of these manipulations’ steps: pulling over a column with the mouse, renaming a row, adding a zero to an empty cell. This makes senior supervisory oversight and external audit very costly.

Our data comes with full history: all changes are visible, and we even open the code or algorithm that processed the raw data. Your analysts can still use their favourite spreadsheet or statistical software application, but they can start from a clean, tidy dataset, with all data wrangling, currency and unit conversion, imputation and other low-priority but important tasks done and logged.

Metadata

Tue, 01 Jun 2021 11:00:00 +0000

Our observatory has a new data API which allows access to our daily refreshing open data. You can access the API via api.economy.dataobservatory.eu (apologies for the ugly, temporary subdomain masking!).

All the data and the metadata are available as open data, without database use restrictions, under the ODbL license. However, the metadata contents are not finalized yet. We are currently working on a solution that applies the FAIR Guiding Principles for scientific data management and stewardship, and fulfills the mandatory requirements of the Dublic Core metadata standards and at the same time the mandatory requirements, and most of the recommended requirements of DataCite. These changes will be effective before 1 July 2021.

The Competition Data Observatory temporarily shares an API with the Economy Data Observatory, which serves as an incubator for similar economy-oriented reproducible research resources.

api.economy.dataobservatory.eu: processing metadata

Descriptive Metadata


Identifier	An unambiguous reference to the resource within a given context. (Dublin Core item), but several identifiders allowed, and we will use several of them.
Creator	The main researchers involved in producing the data, or the authors of the publication, in priority order. To supply multiple creators, repeat this property. (Extends the Dublin Core with multiple authors, and legal persons, and adds affiliation data.)
Title	A name given to the resource. Extends Dublin Core with alternative title, subtitle, translated Title, and other title(s).
Publisher	The name of the entity that holds, archives, publishes prints, distributes, releases, issues, or produces the resource. This property will be used to formulate the citation, so consider the prominence of the role. For software, use Publisher for the code repository. (Dublin Core item.)
Publication Year	The year when the data was or will be made publicly available.
Resource Type	We publish Datasets, Images, Report, and Data Papers. (Dublin Core item with controlled vocabulary.)

Recommended for discovery

The Recommended (R) properties are optional, but strongly recommended for interoperability.


Subject	The topic of the resource. (Dublin Core item.)
Contributor	The institution or person responsible for collecting, managing, distributing, or otherwise contributing to the development of the resource. (Extends the Dublin Core with multiple authors, and legal persons, and adds affiliation data.) When applicable, we add Distributor (of the datasets and images), Contact Person, Data Collector, Data Curator, Data Manager, Hosting Institution, Producer (for images), Project Manager, Researcher, Research Group, Rightsholder, Sponsor, Supervisor
Date	A point or period of time associated with an event in the lifecycle of the resource, besides the Dublin Core minimum we add Collected, Created, Issued, Updated, and if necessary, Withdrawn dates to our datasets.
Related Identifier	An identifier or identifiers other than the primary Identifier applied to the resource being registered.
Rights	We give SPDX License List standards rights description with URLs to the actual license. (Dublin Core item: Rights Management)
Description	Recommended for discovery.(Dublin Core item.)
GeoLocation	Similar to Dublin Core item Coverage

The Subject property: we need to set standard coding schemas for each observatory.
Contributor property:
- DataCurator the curator of the dataset, who sets the mandatory properties.
- DataManager the person who keeps the dataset up-to-date.
- ContactPerson the person who can be contacted for reuse requests or bug reports.
The Date property contains the following dates, which are set automatically by the dataobservatory R package:
- Updated when the dataset was updated;
- EarliestObservation, which the earliest, not backcasted, estimated or imputed observation.
- LatestObservation, which the earliest, not backcasted, estimated or imputed observation.
- UpdatedatSource, when the raw data source was last updated.
The GeoLocation is automatically created by the dataobservatory R package.
The Description property optional elements, and we adopted them as follows for the observatories:
- The Abstract is a short, textual description; we try to automate its creation as much as a possible, but some curatorial input is necessary.
- In the TechnicalInfo sub-field, we record automatically the utils::sessionInfo() for computational reproducability. This is automatically created by the dataobservatory R package.
- In the Other sub-field, we record the keywords for structuring the observatory.

Optional

The Optional (O) properties are optional and provide richer description. For findability they are not so important, but to create a web service, they are essential. In the mandatory and recommended fields, we are following other metadata standards and codelists, but in the optional fields we have to build up our own system for the observatories.


Language	A language of the resource. (Dublin Core item.)
Alternative Identifier	An identifier or identifiers other than the primary Identifier applied to the resource being registered.
Size	We give the CSV, downloadable dataset size in bytes.
Format	We give file format information. We mainly use CSV and JSON, and occasionally rds and SPSS types. (Dublin Core item.)
Version	The version number of the resource.
Rights	We give SPDX License List standards rights description with URLs to the actual license. (Dublin Core item: Rights Management)
Funding Reference	We provide the funding reference information when applicable. This is usually mandatory with public funds.
Related Item	We give information about our observatory partners' related research products, awards, grants (also Dublin Core item as Relation.) We particularly include source information when the dataset is derived from another resource (which is a Dublin Core item.)

In the Language we only use English (eng) at the moment.
By default We do not use the Alternative Identifier property. We will do this when the same dataset will be used in several observatories.
The Size property is measured in bytes for the CSV representation of the dataset. During creations, the software creates a temporary CSV file to check if the dataset has no writing problems, and measures the dataset size.
The Version property needs further work. For a daily re-freshing API we need to find an applicable versioning system.
The Funding reference will contain information for donors, sponsors, and co-financing partners.
Our default setting for Rights is the CC-BY-NC-SA-4.0 license and we provide an URI for the license document.
In the RelatedItem we give information about:
- The original (raw) data source.
- Methodological bibilography reference, when needed.
- The open-source statistical software code that processed the data.

Administrative (Processing) Metadata

Administrative Metadata

Like with diamonds, it is better to know the history of a dataset, too. Our administrative metadata contains codelists that follow the SXDX statistical metadata standards, and similarly strucutred information about the processing history of the dataset.

See for further reference The codebook Class.


Observation Status	SDMX Code list for Observation Status 2.2 (CL_OBS_STATUS), such as actual, missing, imputed, etc. values.
Method	If the value is estimated, we provide modelling information.
Unit	We provide the measurement unit of the data (when applicable.)
Frequency	SDMX Code list for Frequency 2.1 (CL_FREQ) frequency values
Codelist	Euros-SDMX Codelist entries for the observational units, such as sex, etc.
Imputation	SDMX Code list for Frequency 2.1 (CL_IMPUT_METH) imputation values
Estimation	The estimation methodology of data that we calculated, together with citation information and URI to the actual processing code
Related Item	We give information about the software code that processed the data (both Dublin Core and DataCite compliant.)

See an example in the The codebook Class article of the dataobservatory R package.