Analysis of the Potential of Selected Big Data Repositories as Data - - PowerPoint PPT Presentation

NTTS, March2015

Analysis of the Potential of Selected Big Data Repositories as Data Sources for Official Statistics

NTTS 2015 New Techniques and Technologies for Statistics

Petrakos M., Santourian A., Farmakis G., Stavropoulos P., Oikonomopoulou G., Ntakou E. Trampeli A.

NTTS, March 2015

Big data and official statistics

NTTS, March 2015

2

VISULIZE ACTIVITIES WITH TIMELINES!

Identify available data Define questions Challenge Official statistics

Big data

Big data: A potential source for official statistics?

Big data paradigm shift... Can Big data used? Answer Define questions Collect data Answer

NTTS, March 2015

4

Automatic Identification System (AIS)

Vessel traffic Gross weight of goods Gross tonnage of vessels Emissions Vessel statistics can be replaced Better coverage, timeliness, accuracy Reduction of response burden

Target population restrictions Data for vessel characteristics may be missing

NTTS, March 2015

Advertisements for house sales and Rentals

5

Real estate advertisements

What can be produced?

• Housing price index and PPPs related

to rental and

• wner
• ccupied

housing

• Owner occupied housing sub-index of

the HICP

• Coverage errors:

Dwellings not yet built are advertised

• Missing information:

List price stated by the owner instead of the actual sale price

NTTS, March 2015

6

45 MILLION status updates each day 400 MILLION tweets each day

Social media messages

Status updates, tweets Keywords expressing emotions Sentiment analysis Sentiment Index

 Sentiment information  Sentiment index can complement official

statistics

 Coverage issues (location identification,

age, sex)

 Creation of domain-specific thesaurus

Well-being

NTTS, March 2015

7

SWOT

Government financial transparency portal

RELATED OFFICIAL STATISTICS  Government expenditure  Main revenue and expenditure items of the general government  General government total expenditure and revenue A primary source for statistics where coverage is complete

NTTS, March 2015

8

Visa Transaction Data

Related official statistics Household consumption expenditure

more than 1.5 MILLIONS are spent on Visa cards EVERY MINUTE 419 MILLION Visa cards in Europe

Outcome

• Flash estimates about structure and

amount of consumption expenditure

• Similar index is already produced by

Visa

• Robust indication of real consumer

spending trends