ESTP course:

" ANNUAL TO QUARTERLY TO MONTHLY DATA "

Agenda

Location

Eurostat, Luxembourg, 15/17 November 2016, Room A2/45

Course Leader

Dario Buono (Eurostat, Unit B1 Methodology and Corporate Architecture)

Riccardo Gatto (Eurostat, Unit F3, Labour Market)

Lecturers

Dario Buono (Eurostat, Unit B1 Methodology and Corporate Architecture)

Riccardo Gatto (Eurostat, Unit F3, Labour Market)

Francesca Di Iorio (Università di Napoli Federico II)

Logistics

Elwira NAVARRE(Eurostat, Unit A2 Learning and Development)

Required Reading

Eurostat (2013), Handbook of Quarterly National Accounts, European

Commission, Luxembourg

Suggested Readings

Dagum E.B., Cholette P. (2006), Benchmarking, Temporal

Distribution, and Reconciliation Methods for Time Series, Springer,

New York.

15 of November 2016, Tuesday

9.30-9.45

Presentation of the course, tour de table (by NAVARRE, Buono)

9.45-11.00

Brief review of temporal disaggregation, balancing and statistical reconciliation of systems of time series (by Buono):

• Flow, index and stock series
• Deriving high frequency data with and without proxies
• Temporal and Accounting constraints

11.00-11.15

coffee-break

11.15-12.30

Balancing a table with fixed marginal totals: RAS and Stone’s Least Squares adjustment. Dealing with negative entries: RAS-PM. Procedures without indicators: Boot, Feibes and Lisman (BFL). The movement preservation principle: modified Denton PFD benchmarking. (by Gatto)

12.30-14.00

Lunch break

14.00-15.15

Practical exercise: getting familiar with the tool (by Di Iorio)

15.15-15.30

coffee-break

15.30-16.30

Practical exercise: first results (by Buono)

16.30-17.00

Question time for face-to face discussion with the trainers

16 of November 2016, Wednesday

9.30-11.00

The statistical reconciliation of systems of time series; main field of application: Quarterly National Accounts, Labour Force, Industrial Production Indices, (by Buono)

11.00-11.15

coffee-break

11.15-12.30

Regression-based temporal disaggregation: Two-step adjustment procedures. Optimal regression-based temporal disaggregation (by Gatto)

12.30-14.00

Lunch break

14.00-15.15

Practical exercise: full results (by Di Iorio)

15.15-15.30

coffee-break

15.30-16.15

Practical exercise: full results (by Di Iorio)

16.15-16.30

Question time for face-to face discussion with the trainers

17 of November 2016, Thursday

9.30-11.00

Optimal regression-based temporal disaggregation techniques: Chow-Lin, Fernandez and Litterman. (by Gatto)

11.00-11.15

coffee-break

11.15-12.30

Temporal and contemporaneous aggregation constraints.

Two-step reconciliation techniques of preliminary time series: 2S-QR and 2S-ST procedures. (by Gatto and by Di Iorio)

12.30-14.00

Lunch break

14.00-15.30

Show and tell with your data set (by the participants!!)

15.30-15.45

coffee-break

15.45-16.30

Conclusions and evaluation of the course