1 / 16

Anna Ciammola and Donatella Tuzi ISTAT - Italy

Internal Coherence in Seasonally Adjusted Chain Laspeyres Indices An Application to the Italian Hourly Labour Cost Indicators. Anna Ciammola and Donatella Tuzi ISTAT - Italy. European Conference on Quality in Official Statistics Q2010. Helsinki, 4-6 May 2010. Layout.

zoe-bowen
Télécharger la présentation

Anna Ciammola and Donatella Tuzi ISTAT - Italy

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Internal Coherence in Seasonally Adjusted Chain Laspeyres Indices An Application to the Italian Hourly Labour Cost Indicators Anna Ciammola and Donatella Tuzi ISTAT - Italy European Conference on Quality in Official Statistics Q2010 Helsinki, 4-6 May 2010

  2. Layout • Objectives of the presentation • The problem: Seasonal Adjustment (SA) of Italian hourly Labour Cost Indicators (LCI) • The solution: a proposal to aggregate chain linked Laspeyres indices • Results and conclusions

  3. Objectives • Internal coherence in a system of SA time series • Number of components  few or many • Approach for SA  direct or indirect • Implemention of the indirect approach for chain Laspeyres indices • Chain linking non-additivity • A proposal to “restore” additivity

  4. The problem (1) The Italian hourly LCI system (EC 450/2003) Wages Other Costs Total Cost B C D . . . L M N Elementary indices Chain Laspeyres indices B-N

  5. The problem (2) Seasonal adjustment of the LCI system (a) Direct approach Independent treatment of wages, other costs andtotal cost total cost < min (wages, other costs) total cost > max (wages, other costs) Internal coherence not fulfilled (more evident to users for period-on-period changes)

  6. The problem (3) Seasonal adjustment of the LCI system (b) Indirect approach total cost = f1(wages, other costs)  section B-N = f2(sections)  component f1andf2 ~ weighted average Unknowns of the problem  weights of f1 and f2 Internal coherence always fulfilled

  7. The proposal: sectorial total cost (sC) • Elementary indices • Proposal: indices as weighted averages

  8. The proposal: B-N chain indices (Sc) The starting point definitions • Laspeyres indicesin the previous year base(a-1) • Chain linked indices in the fixed base (b) • LSc, l, l+1 annual average of quarterly l LCISc, tl+1 • Chain linking non-additivity • Weights unsuited to the indirect approach

  9. The proposal: B-N chain indices (Sc) A new weighting system • Indirect approach for seasonal adjustment • f  weighted average • Weights to “restore” additivity in the LCI system • if

  10. Results (1) B-N total cost - direct and indirect approach (a)

  11. Results (2) B-N total cost - direct and indirect approach (b)

  12. Results (3) Assessment of the quality of seasonal adjustment • Residual seasonality • Smoothness measures • Stability • Sliding spans • Revisions history

  13. Conclusions • Direct and indirect approach almost equivalent in terms of residual seasonality and smoothness • Sliding spans computable only for some NACE sections • Indirect approach slightly outperforms the indirect one in terms of revisions history Internal coherence as crucial criterion in the choice of the indirect approach

  14. Thank you!

  15. T1 - Incoherencies in the LCI system Number of incoherencies on q-on-q changes (2000Q2-2009Q4) NACE Rev. 2 Sections – Total cost aggregate F (9.8%) 16 41.0% H (9.0%) 6 15.4% J (5.2%) 12 30.8% M (4.3%) 7 17.9% Labour cost components – B-N aggregate Wages 0 0% Other costs 0 0% Total cost 0 0%

  16. T2 – Revisions history Mean absolute differences on q-on-q changes Labour cost components – B-N aggregate Direct approach 1 step 2 steps 3steps 4steps Wages 0.13 0.21 0.36 0.67 Other costs 0.09 0.21 0.30 0.54 Total cost 0.14 0.24 0.39 0.68 Indirect approach 1 step 2 steps 3steps 4steps Wages 0.12 0.12 0.13 0.25 Other costs 0.07 0.06 0.10 0.26 Total cost 0.10 0.10 0.12 0.25

More Related