SlideShare a Scribd company logo
1 of 25
Riku Salonen
Regression composite estimation for the
Finnish LFS from a practical perspective
May 15 - 16, 2014 2LFS Workshop in Rome
Outline
 Design of the FI-LFS
 The idea of RC-estimator
 Empirical results
 Conclusions and future work
May 15 - 16, 2014 3LFS Workshop in Rome
The FI-LFS
 Monthly survey on individuals of the age 15-74
 Sample size is 12 500 divided into 5 waves
 It provides monthly, quarterly and annual results
 Sampling design is stratified systematic sampling
 The strata: Mainland Finland and Åland Islands
 In both stratum systematic random selection is applied
to the frame sorted according to the domicile code
Implicit geographic stratification
May 15 - 16, 2014 4LFS Workshop in Rome
Rotation panel design
 Partially overlapping samples
 Each sample person is in sample 5 times during 15 months
 The monthly rotation pattern: 1-(2)-1-(2)-1-(5)-1-(2)-1
 No month to month overlap
 60% quarter to quarter theoretical overlap
 40% year to year theoretical overlap
 Independence: monthly samples in each three-month
period quarterly sample consists of separate monthly
samples
May 15 - 16, 2014 5LFS Workshop in Rome
Sample allocation (1)
 The half-year sample is drawn two times a year
 It is allocated into six equal part - one for the next six months
The half-year sample
(e.g. Jan-June 2014)
Jan
Mar
Apr
May
June
The monthly sample
(e.g. Jan 2014)
Wave (1)
Wave (2)
Wave (3)
Wave (4)
Wave (5)
”Sample bank”
Earlier
samples
Feb
May 15 - 16, 2014 6LFS Workshop in Rome
Sample allocation (2)
 The monthly sample is
i) divided into five waves
 wave (1) come from the half-year sample
 waves (2) to (5) come from ”sample bank”
ii) distributed uniformly across the weeks of the month
(4 or 5 reference weeks)
 The quarterly sample (usually 13 reference weeks) consist
of three separate and independent monthly samples.
May 15 - 16, 2014 7LFS Workshop in Rome
Weighting procedure
 The weighting procedure (GREG estimator) of the FI-LFS
on monthly level is whole based on quarterly ja annual
weighting also.
 For this purpose
i) the monthly weights need to be divided by three to
create quarterly weights and
ii) the monthly weights need to be divided by twelve to
create annual weights.
 This automatically means that monthly, quarterly and
annual estimates are consistent.
May 15 - 16, 2014 8LFS Workshop in Rome
The idea of RC-estimator
 Extends the current GREG estimator used FI-LFS.
 To improve the estimate by incorporating information from
previous wave (or waves) of interview.
 Takes the advantage of correlations over time.
May 15 - 16, 2014 9LFS Workshop in Rome
RC estimation procedure
 The technical details and formulas of the RC estimation
method with application to the FI-LFS are summarized in
the workshop paper and in Salonen (2007).
 RC estimator introduced by Singh et. al, Fuller et. al and
Gambino et. al (2001).
 Examined further by Bocci and Beaumont (2005).
May 15 - 16, 2014 10LFS Workshop in Rome
RC estimation system implementation
 The RC estimator can be implemented within the FI-LFS
estimation system by adding control totals and auxiliary
variables to the estimation program.
 It can be performed by using, with minor modification,
standard software for GREG estimation, such as ETOS.
 It yields a single set of estimation weights.
May 15 - 16, 2014 11LFS Workshop in Rome
Control totals of auxiliary variables
 Population control totals
 Assumed to be population values
 Composite control totals
 Estimated control totals
May 15 - 16, 2014 12LFS Workshop in Rome
Population control totals
 Population totals taken from administrative registers
 sex (2)
 age (12)
 region (20)
 employment status in Ministry of Labour's job-seeker
register (8)
 Obs! Weekly balancing of weights on monthly level is also
included in the calibration (4 or 5 reference weeks).
May 15 - 16, 2014 13LFS Workshop in Rome
Composite control totals
 Composite control totals are estimates from the previous
wave of interview
 Employed and unemployed by age/sex groups (8)
 Employed and unemployed by NUTS2 (8)
 Employment by Standard Industrial Classification (7)
May 15 - 16, 2014 14LFS Workshop in Rome
Table 1. Population and composite control totals for RC estimation
var N mar1 mar2 … mar20
region 20 282 759 345 671 … 786 285
sex 2 1 995 190 1 994 188 …
age group 12 330 875 328 105 …
reference week (4 or 5) 4 997 346 997 346 …
register-based job-seeker status 8 68 429 115 171 …
Z_emp1 0 161 128
:
Z_emp4 0 1 050 431
Z_une1 0 17 846
: COMPOSITE
Z_une4 0 67 283 CONTROL
Z_nace1 0 115 624 TOTALS
:
Z_nace7 0 804 126
Z_nuts1 0 1 338 271
:
Z_nuts8 0 22 555

May 15 - 16, 2014 15LFS Workshop in Rome
Composite auxiliary variables
 Overlapping part of the sample
 Variables are taken from the previous wave of interview
 Non-overlapping part of the sample
 The values of variables are imputed
May 15 - 16, 2014 16LFS Workshop in Rome
Example 1. Overlapping
January 2014
Wave 1
Wave 2
Wave 3
Wave 4
Wave 5
Previous interview
none
October 2013
October 2013
(July 2013)
October 2013
Dependence: Theoretical overlap wave-to-wave is 4/5 (80%)
May 15 - 16, 2014 17LFS Workshop in Rome
Empirical results
 We have compared the RC estimator to the GREG
estimator in the FI-LFS real data (2006-2010)
 Here we have used the ETOS program for point and
variance estimation (Taylor linearisation method).
 Relative efficiency (RE) can be formulated as
 A value of RE greater than 100 indicates that the RC
estimator is more efficient than the GREG estimator.
 
 yrc
ygr
tV
tV
RE
ˆˆ
100ˆˆ 

May 15 - 16, 2014 18LFS Workshop in Rome
Table 2. Distribution of calibrated weights for GREG and RC estimators
(e.g 2nd quarter of 2006)
The calibrated weights are obtained by the ETOS program. The results
show that the variation of the RC weights is smaller than that of the
GREG weights.
GREG RCStatistics for
calibrated weights
Minimum 42.81 50.66
Maximum 416.29 221.87
Average 137.69 137.69
Median 135.12 137.15
May 15 - 16, 2014 19LFS Workshop in Rome
Table 3. Relative efficiency (RE, %) of estimates for the quarterly level of
employment and unemployment by sex
Quarterly level estimates
Labour force status Sex RE (%)
Employed Male 184,8
Female 178,6
Both sexes 173,5
Unemployed Male 114,2
Female 110,3
Both sexes 106,4
May 15 - 16, 2014 20LFS Workshop in Rome
Table 4. Relative efficiency (RE, %) of estimates for the monthly level of
employment and unemployment by industrial classification
Monthly level estimates
NACE Sample size RE (%)
Agriculture 251 395,6
Manufacturing 1 075 461,4
Construction 375 373,6
Wholesale and retail trade 899 318,3
Transport, storage and communication 390 365,4
Financial intermediation 802 398,3
Public administration 1 865 361,1
May 15 - 16, 2014 21LFS Workshop in Rome
Table 5. Relative efficiency (RE, %) of estimates for the quarterly level of
employment and unemployment by industrial classification
Quarterly level estimates
NACE Sample size RE (%)
Agriculture 788 358,2
Manufacturing 3 287 406,6
Construction 1 098 375,7
Wholesale and retail trade 2 698 375,6
Transport, storage and communication 1 216 369,1
Financial intermediation 2 403 370,8
Public administration 5 951 361,3
May 15 - 16, 2014 22LFS Workshop in Rome
Conclusions (1)
 For the variables that were included as composite control
totals, there are substantial gains in efficiency for estimates
 For some variables it is future possible to publish monthly
estimates where only quarterly estimates are published
now?
 Leading to internal consistency of estimates
 Employment + Unemployment = Labour Force
 Labour Force + Not In Labour Force = Population 15 to 74
May 15 - 16, 2014 23LFS Workshop in Rome
Conclusions (2)
 It can be performed by using, with minor modification,
standard software for GREG estimation, such as ETOS
 It yields a single set of estimation weights
 The results are well comparable with results reported from
other countries
 Chen and Liu (2002): the Canadian LFS
 Bell (2001): the Australian LFS
May 15 - 16, 2014 24LFS Workshop in Rome
Future work
 Analysis of potential imputation methods for the non-
overlapping part of the sample?
 Analysis of alternative variance estimators (Dever and
Valliant, 2010)?
 Incorporating information from all potential previous waves
of interview
May 15 - 16, 2014 25LFS Workshop in Rome
MAIN REFERENCES
BEAUMONT, J.-F. and BOCCI, C. (2005). A Refinement of the Regression Composite
Estimator in the Labour Force Survey for Change Estimates. SSC Annual Meeting,
Proceedings of the Survey Methods Section, June 2005.
CHEN, E.J. and LIU, T.P. (2002). Choices of Alpha Value in Regression Composite
Estimation for the Canadian Labour Force Survey: Impacts and Evaluation. Methodology
Branch Working Paper, HSMD-2002-005E, Statistics Canada.
DEVER, A.D., and VALLIANT, R. (2010). A Comparison of Variance Estimators for
Poststratification to Estimated Control Totals. Survey Methodology, 36, 45-56.
FULLER, W.A., and RAO, J.N.K. (2001). A Regression Composite Estimator with
Application to the Canadian Labour Force Survey. Survey Methodology, 27, 45-51.
GAMBINO, J., KENNEDY, B., and SINGH, M.P. (2001). Regression Composite Estimation
for the Canadian Labour Force Survey: Evaluation ja Implementation. Survey
Methodology, 27, 65-74.
SALONEN, R. (2007). Regression Composite Estimation with Application to the Finnish
Labour Force Survey. Statistics in Transition, 8, 503-517.
SINGH, A.C., KENNEDY, B., and WU, S. (2001). Regression Composite Estimation for the
Canadian Labour Force Survey with a Rotating Panel Design. Survey Methodology, 27,
33-44.

More Related Content

Similar to R. Salonen - Regression composite estimation for the Finnish LFS from a practical perspective

Iwsm2014 an evaluation of simple function point as a replacement of ifpug f...
Iwsm2014   an evaluation of simple function point as a replacement of ifpug f...Iwsm2014   an evaluation of simple function point as a replacement of ifpug f...
Iwsm2014 an evaluation of simple function point as a replacement of ifpug f...
Nesma
 
Vocational Rehabilitation and Employment Longitudenal Study Report to Congress
Vocational Rehabilitation and Employment Longitudenal Study Report to CongressVocational Rehabilitation and Employment Longitudenal Study Report to Congress
Vocational Rehabilitation and Employment Longitudenal Study Report to Congress
Rob Wilson
 
(BR) JK Monitoring and Evaluation 02 August 2016
(BR) JK Monitoring and Evaluation 02 August 2016(BR) JK Monitoring and Evaluation 02 August 2016
(BR) JK Monitoring and Evaluation 02 August 2016
Joyce Khunou
 
SCOR for Offices 2015
SCOR for Offices 2015SCOR for Offices 2015
SCOR for Offices 2015
Olisa Pal
 
SENSITIVITY ANALYSIS OF INFORMATION RETRIEVAL METRICS
SENSITIVITY ANALYSIS OF INFORMATION RETRIEVAL METRICS SENSITIVITY ANALYSIS OF INFORMATION RETRIEVAL METRICS
SENSITIVITY ANALYSIS OF INFORMATION RETRIEVAL METRICS
ijcsit
 
Iwsm2014 the effect of highlighting error categories in fsm training on the...
Iwsm2014   the effect of highlighting error categories in fsm training on the...Iwsm2014   the effect of highlighting error categories in fsm training on the...
Iwsm2014 the effect of highlighting error categories in fsm training on the...
Nesma
 

Similar to R. Salonen - Regression composite estimation for the Finnish LFS from a practical perspective (20)

L. Di Consiglio, S. Loriga, A. Martini, R. Ranaldi - The Italian LFS sampling...
L. Di Consiglio, S. Loriga, A. Martini, R. Ranaldi - The Italian LFS sampling...L. Di Consiglio, S. Loriga, A. Martini, R. Ranaldi - The Italian LFS sampling...
L. Di Consiglio, S. Loriga, A. Martini, R. Ranaldi - The Italian LFS sampling...
 
M. Juillard S. Le Minezh - ow is the uniform distribution of reference weeks ...
M. Juillard S. Le Minezh - ow is the uniform distribution of reference weeks ...M. Juillard S. Le Minezh - ow is the uniform distribution of reference weeks ...
M. Juillard S. Le Minezh - ow is the uniform distribution of reference weeks ...
 
Explaining differences in efficiency: the case of local government literature
Explaining differences in efficiency: the case of local government literature Explaining differences in efficiency: the case of local government literature
Explaining differences in efficiency: the case of local government literature
 
CGIAR Consortium/System Office - Monitoring, Evaluation and Learning
CGIAR Consortium/System Office - Monitoring, Evaluation and Learning CGIAR Consortium/System Office - Monitoring, Evaluation and Learning
CGIAR Consortium/System Office - Monitoring, Evaluation and Learning
 
currency wars
currency warscurrency wars
currency wars
 
Iwsm2014 an evaluation of simple function point as a replacement of ifpug f...
Iwsm2014   an evaluation of simple function point as a replacement of ifpug f...Iwsm2014   an evaluation of simple function point as a replacement of ifpug f...
Iwsm2014 an evaluation of simple function point as a replacement of ifpug f...
 
The Effect Of Economic Value Added And Earning Per Share To Stocks Return (Pa...
The Effect Of Economic Value Added And Earning Per Share To Stocks Return (Pa...The Effect Of Economic Value Added And Earning Per Share To Stocks Return (Pa...
The Effect Of Economic Value Added And Earning Per Share To Stocks Return (Pa...
 
Vocational Rehabilitation and Employment Longitudenal Study Report to Congress
Vocational Rehabilitation and Employment Longitudenal Study Report to CongressVocational Rehabilitation and Employment Longitudenal Study Report to Congress
Vocational Rehabilitation and Employment Longitudenal Study Report to Congress
 
(BR) JK Monitoring and Evaluation 02 August 2016
(BR) JK Monitoring and Evaluation 02 August 2016(BR) JK Monitoring and Evaluation 02 August 2016
(BR) JK Monitoring and Evaluation 02 August 2016
 
Exchange rate regimes and macroeconomic perfomance zithe p. machewere
Exchange rate regimes and macroeconomic perfomance  zithe p. machewereExchange rate regimes and macroeconomic perfomance  zithe p. machewere
Exchange rate regimes and macroeconomic perfomance zithe p. machewere
 
SCOR for Offices 2015
SCOR for Offices 2015SCOR for Offices 2015
SCOR for Offices 2015
 
SENSITIVITY ANALYSIS OF INFORMATION RETRIEVAL METRICS
SENSITIVITY ANALYSIS OF INFORMATION RETRIEVAL METRICS SENSITIVITY ANALYSIS OF INFORMATION RETRIEVAL METRICS
SENSITIVITY ANALYSIS OF INFORMATION RETRIEVAL METRICS
 
Iwsm2014 the effect of highlighting error categories in fsm training on the...
Iwsm2014   the effect of highlighting error categories in fsm training on the...Iwsm2014   the effect of highlighting error categories in fsm training on the...
Iwsm2014 the effect of highlighting error categories in fsm training on the...
 
Predictive Analytics and Strategic Enrolment Management
Predictive Analytics and Strategic Enrolment ManagementPredictive Analytics and Strategic Enrolment Management
Predictive Analytics and Strategic Enrolment Management
 
Idh q3 2014 summary report final
Idh q3 2014 summary report finalIdh q3 2014 summary report final
Idh q3 2014 summary report final
 
K. Baumgartner, A. Meraner, A. Kowarik - Longitudinal Weights for the Product...
K. Baumgartner, A. Meraner, A. Kowarik - Longitudinal Weights for the Product...K. Baumgartner, A. Meraner, A. Kowarik - Longitudinal Weights for the Product...
K. Baumgartner, A. Meraner, A. Kowarik - Longitudinal Weights for the Product...
 
Applying Statistical Forecasting Of Project Duration To Earned Schedule-Longe...
Applying Statistical Forecasting Of Project Duration To Earned Schedule-Longe...Applying Statistical Forecasting Of Project Duration To Earned Schedule-Longe...
Applying Statistical Forecasting Of Project Duration To Earned Schedule-Longe...
 
Services in EU what kinds of regulatory policy enhance productivity van der M...
Services in EU what kinds of regulatory policy enhance productivity van der M...Services in EU what kinds of regulatory policy enhance productivity van der M...
Services in EU what kinds of regulatory policy enhance productivity van der M...
 
Sampling design issues in Italian experience on scanner data and the possible...
Sampling design issues in Italian experience on scanner data and the possible...Sampling design issues in Italian experience on scanner data and the possible...
Sampling design issues in Italian experience on scanner data and the possible...
 
A Bug Report Analysis and Search Tool (presentation for M.Sc. degree)
A Bug Report Analysis and Search Tool (presentation for M.Sc. degree)A Bug Report Analysis and Search Tool (presentation for M.Sc. degree)
A Bug Report Analysis and Search Tool (presentation for M.Sc. degree)
 

More from Istituto nazionale di statistica

More from Istituto nazionale di statistica (20)

Censimenti Permanenti Istituzioni non profit
Censimenti Permanenti Istituzioni non profitCensimenti Permanenti Istituzioni non profit
Censimenti Permanenti Istituzioni non profit
 
Censimenti Permanenti Istituzioni non profit
Censimenti Permanenti Istituzioni non profitCensimenti Permanenti Istituzioni non profit
Censimenti Permanenti Istituzioni non profit
 
Censimenti Permanenti Istituzioni non profit
Censimenti Permanenti Istituzioni non profitCensimenti Permanenti Istituzioni non profit
Censimenti Permanenti Istituzioni non profit
 
Censimenti Permanenti Istituzioni non profit
Censimenti Permanenti Istituzioni non profitCensimenti Permanenti Istituzioni non profit
Censimenti Permanenti Istituzioni non profit
 
Censimenti Permanenti Istituzioni non profit
Censimenti Permanenti Istituzioni non profitCensimenti Permanenti Istituzioni non profit
Censimenti Permanenti Istituzioni non profit
 
Censimenti Permanenti Istituzioni non profit
Censimenti Permanenti Istituzioni non profitCensimenti Permanenti Istituzioni non profit
Censimenti Permanenti Istituzioni non profit
 
Censimento Permanente Istituzioni Pubbliche
Censimento Permanente Istituzioni PubblicheCensimento Permanente Istituzioni Pubbliche
Censimento Permanente Istituzioni Pubbliche
 
Censimento Permanente Istituzioni Pubbliche
Censimento Permanente Istituzioni PubblicheCensimento Permanente Istituzioni Pubbliche
Censimento Permanente Istituzioni Pubbliche
 
Censimento Permanente Istituzioni Pubbliche
Censimento Permanente Istituzioni PubblicheCensimento Permanente Istituzioni Pubbliche
Censimento Permanente Istituzioni Pubbliche
 
Censimento Permanente Istituzioni Pubbliche
Censimento Permanente Istituzioni PubblicheCensimento Permanente Istituzioni Pubbliche
Censimento Permanente Istituzioni Pubbliche
 
14a Conferenza Nazionale di Statisticacnstatistica14
14a Conferenza Nazionale di Statisticacnstatistica1414a Conferenza Nazionale di Statisticacnstatistica14
14a Conferenza Nazionale di Statisticacnstatistica14
 
14a Conferenza Nazionale di Statistica
14a Conferenza Nazionale di Statistica14a Conferenza Nazionale di Statistica
14a Conferenza Nazionale di Statistica
 
14a Conferenza Nazionale di Statistica
14a Conferenza Nazionale di Statistica14a Conferenza Nazionale di Statistica
14a Conferenza Nazionale di Statistica
 
14a Conferenza Nazionale di Statistica
14a Conferenza Nazionale di Statistica14a Conferenza Nazionale di Statistica
14a Conferenza Nazionale di Statistica
 
14a Conferenza Nazionale di Statistica
14a Conferenza Nazionale di Statistica14a Conferenza Nazionale di Statistica
14a Conferenza Nazionale di Statistica
 
14a Conferenza Nazionale di Statistica
14a Conferenza Nazionale di Statistica14a Conferenza Nazionale di Statistica
14a Conferenza Nazionale di Statistica
 
14a Conferenza Nazionale di Statistica
14a Conferenza Nazionale di Statistica14a Conferenza Nazionale di Statistica
14a Conferenza Nazionale di Statistica
 
14a Conferenza Nazionale di Statistica
14a Conferenza Nazionale di Statistica14a Conferenza Nazionale di Statistica
14a Conferenza Nazionale di Statistica
 
14a Conferenza Nazionale di Statistica
14a Conferenza Nazionale di Statistica14a Conferenza Nazionale di Statistica
14a Conferenza Nazionale di Statistica
 
14a Conferenza Nazionale di Statistica
14a Conferenza Nazionale di Statistica14a Conferenza Nazionale di Statistica
14a Conferenza Nazionale di Statistica
 

Recently uploaded

Seal of Good Local Governance (SGLG) 2024Final.pptx
Seal of Good Local Governance (SGLG) 2024Final.pptxSeal of Good Local Governance (SGLG) 2024Final.pptx
Seal of Good Local Governance (SGLG) 2024Final.pptx
negromaestrong
 
1029-Danh muc Sach Giao Khoa khoi 6.pdf
1029-Danh muc Sach Giao Khoa khoi  6.pdf1029-Danh muc Sach Giao Khoa khoi  6.pdf
1029-Danh muc Sach Giao Khoa khoi 6.pdf
QucHHunhnh
 
Activity 01 - Artificial Culture (1).pdf
Activity 01 - Artificial Culture (1).pdfActivity 01 - Artificial Culture (1).pdf
Activity 01 - Artificial Culture (1).pdf
ciinovamais
 

Recently uploaded (20)

Asian American Pacific Islander Month DDSD 2024.pptx
Asian American Pacific Islander Month DDSD 2024.pptxAsian American Pacific Islander Month DDSD 2024.pptx
Asian American Pacific Islander Month DDSD 2024.pptx
 
Explore beautiful and ugly buildings. Mathematics helps us create beautiful d...
Explore beautiful and ugly buildings. Mathematics helps us create beautiful d...Explore beautiful and ugly buildings. Mathematics helps us create beautiful d...
Explore beautiful and ugly buildings. Mathematics helps us create beautiful d...
 
SOC 101 Demonstration of Learning Presentation
SOC 101 Demonstration of Learning PresentationSOC 101 Demonstration of Learning Presentation
SOC 101 Demonstration of Learning Presentation
 
Seal of Good Local Governance (SGLG) 2024Final.pptx
Seal of Good Local Governance (SGLG) 2024Final.pptxSeal of Good Local Governance (SGLG) 2024Final.pptx
Seal of Good Local Governance (SGLG) 2024Final.pptx
 
2024-NATIONAL-LEARNING-CAMP-AND-OTHER.pptx
2024-NATIONAL-LEARNING-CAMP-AND-OTHER.pptx2024-NATIONAL-LEARNING-CAMP-AND-OTHER.pptx
2024-NATIONAL-LEARNING-CAMP-AND-OTHER.pptx
 
ComPTIA Overview | Comptia Security+ Book SY0-701
ComPTIA Overview | Comptia Security+ Book SY0-701ComPTIA Overview | Comptia Security+ Book SY0-701
ComPTIA Overview | Comptia Security+ Book SY0-701
 
Application orientated numerical on hev.ppt
Application orientated numerical on hev.pptApplication orientated numerical on hev.ppt
Application orientated numerical on hev.ppt
 
Mehran University Newsletter Vol-X, Issue-I, 2024
Mehran University Newsletter Vol-X, Issue-I, 2024Mehran University Newsletter Vol-X, Issue-I, 2024
Mehran University Newsletter Vol-X, Issue-I, 2024
 
Python Notes for mca i year students osmania university.docx
Python Notes for mca i year students osmania university.docxPython Notes for mca i year students osmania university.docx
Python Notes for mca i year students osmania university.docx
 
1029-Danh muc Sach Giao Khoa khoi 6.pdf
1029-Danh muc Sach Giao Khoa khoi  6.pdf1029-Danh muc Sach Giao Khoa khoi  6.pdf
1029-Danh muc Sach Giao Khoa khoi 6.pdf
 
Dyslexia AI Workshop for Slideshare.pptx
Dyslexia AI Workshop for Slideshare.pptxDyslexia AI Workshop for Slideshare.pptx
Dyslexia AI Workshop for Slideshare.pptx
 
Activity 01 - Artificial Culture (1).pdf
Activity 01 - Artificial Culture (1).pdfActivity 01 - Artificial Culture (1).pdf
Activity 01 - Artificial Culture (1).pdf
 
Accessible Digital Futures project (20/03/2024)
Accessible Digital Futures project (20/03/2024)Accessible Digital Futures project (20/03/2024)
Accessible Digital Futures project (20/03/2024)
 
Spatium Project Simulation student brief
Spatium Project Simulation student briefSpatium Project Simulation student brief
Spatium Project Simulation student brief
 
UGC NET Paper 1 Mathematical Reasoning & Aptitude.pdf
UGC NET Paper 1 Mathematical Reasoning & Aptitude.pdfUGC NET Paper 1 Mathematical Reasoning & Aptitude.pdf
UGC NET Paper 1 Mathematical Reasoning & Aptitude.pdf
 
Introduction to Nonprofit Accounting: The Basics
Introduction to Nonprofit Accounting: The BasicsIntroduction to Nonprofit Accounting: The Basics
Introduction to Nonprofit Accounting: The Basics
 
How to Create and Manage Wizard in Odoo 17
How to Create and Manage Wizard in Odoo 17How to Create and Manage Wizard in Odoo 17
How to Create and Manage Wizard in Odoo 17
 
PROCESS RECORDING FORMAT.docx
PROCESS      RECORDING        FORMAT.docxPROCESS      RECORDING        FORMAT.docx
PROCESS RECORDING FORMAT.docx
 
TỔNG ÔN TẬP THI VÀO LỚP 10 MÔN TIẾNG ANH NĂM HỌC 2023 - 2024 CÓ ĐÁP ÁN (NGỮ Â...
TỔNG ÔN TẬP THI VÀO LỚP 10 MÔN TIẾNG ANH NĂM HỌC 2023 - 2024 CÓ ĐÁP ÁN (NGỮ Â...TỔNG ÔN TẬP THI VÀO LỚP 10 MÔN TIẾNG ANH NĂM HỌC 2023 - 2024 CÓ ĐÁP ÁN (NGỮ Â...
TỔNG ÔN TẬP THI VÀO LỚP 10 MÔN TIẾNG ANH NĂM HỌC 2023 - 2024 CÓ ĐÁP ÁN (NGỮ Â...
 
Unit-V; Pricing (Pharma Marketing Management).pptx
Unit-V; Pricing (Pharma Marketing Management).pptxUnit-V; Pricing (Pharma Marketing Management).pptx
Unit-V; Pricing (Pharma Marketing Management).pptx
 

R. Salonen - Regression composite estimation for the Finnish LFS from a practical perspective

  • 1. Riku Salonen Regression composite estimation for the Finnish LFS from a practical perspective
  • 2. May 15 - 16, 2014 2LFS Workshop in Rome Outline  Design of the FI-LFS  The idea of RC-estimator  Empirical results  Conclusions and future work
  • 3. May 15 - 16, 2014 3LFS Workshop in Rome The FI-LFS  Monthly survey on individuals of the age 15-74  Sample size is 12 500 divided into 5 waves  It provides monthly, quarterly and annual results  Sampling design is stratified systematic sampling  The strata: Mainland Finland and Åland Islands  In both stratum systematic random selection is applied to the frame sorted according to the domicile code Implicit geographic stratification
  • 4. May 15 - 16, 2014 4LFS Workshop in Rome Rotation panel design  Partially overlapping samples  Each sample person is in sample 5 times during 15 months  The monthly rotation pattern: 1-(2)-1-(2)-1-(5)-1-(2)-1  No month to month overlap  60% quarter to quarter theoretical overlap  40% year to year theoretical overlap  Independence: monthly samples in each three-month period quarterly sample consists of separate monthly samples
  • 5. May 15 - 16, 2014 5LFS Workshop in Rome Sample allocation (1)  The half-year sample is drawn two times a year  It is allocated into six equal part - one for the next six months The half-year sample (e.g. Jan-June 2014) Jan Mar Apr May June The monthly sample (e.g. Jan 2014) Wave (1) Wave (2) Wave (3) Wave (4) Wave (5) ”Sample bank” Earlier samples Feb
  • 6. May 15 - 16, 2014 6LFS Workshop in Rome Sample allocation (2)  The monthly sample is i) divided into five waves  wave (1) come from the half-year sample  waves (2) to (5) come from ”sample bank” ii) distributed uniformly across the weeks of the month (4 or 5 reference weeks)  The quarterly sample (usually 13 reference weeks) consist of three separate and independent monthly samples.
  • 7. May 15 - 16, 2014 7LFS Workshop in Rome Weighting procedure  The weighting procedure (GREG estimator) of the FI-LFS on monthly level is whole based on quarterly ja annual weighting also.  For this purpose i) the monthly weights need to be divided by three to create quarterly weights and ii) the monthly weights need to be divided by twelve to create annual weights.  This automatically means that monthly, quarterly and annual estimates are consistent.
  • 8. May 15 - 16, 2014 8LFS Workshop in Rome The idea of RC-estimator  Extends the current GREG estimator used FI-LFS.  To improve the estimate by incorporating information from previous wave (or waves) of interview.  Takes the advantage of correlations over time.
  • 9. May 15 - 16, 2014 9LFS Workshop in Rome RC estimation procedure  The technical details and formulas of the RC estimation method with application to the FI-LFS are summarized in the workshop paper and in Salonen (2007).  RC estimator introduced by Singh et. al, Fuller et. al and Gambino et. al (2001).  Examined further by Bocci and Beaumont (2005).
  • 10. May 15 - 16, 2014 10LFS Workshop in Rome RC estimation system implementation  The RC estimator can be implemented within the FI-LFS estimation system by adding control totals and auxiliary variables to the estimation program.  It can be performed by using, with minor modification, standard software for GREG estimation, such as ETOS.  It yields a single set of estimation weights.
  • 11. May 15 - 16, 2014 11LFS Workshop in Rome Control totals of auxiliary variables  Population control totals  Assumed to be population values  Composite control totals  Estimated control totals
  • 12. May 15 - 16, 2014 12LFS Workshop in Rome Population control totals  Population totals taken from administrative registers  sex (2)  age (12)  region (20)  employment status in Ministry of Labour's job-seeker register (8)  Obs! Weekly balancing of weights on monthly level is also included in the calibration (4 or 5 reference weeks).
  • 13. May 15 - 16, 2014 13LFS Workshop in Rome Composite control totals  Composite control totals are estimates from the previous wave of interview  Employed and unemployed by age/sex groups (8)  Employed and unemployed by NUTS2 (8)  Employment by Standard Industrial Classification (7)
  • 14. May 15 - 16, 2014 14LFS Workshop in Rome Table 1. Population and composite control totals for RC estimation var N mar1 mar2 … mar20 region 20 282 759 345 671 … 786 285 sex 2 1 995 190 1 994 188 … age group 12 330 875 328 105 … reference week (4 or 5) 4 997 346 997 346 … register-based job-seeker status 8 68 429 115 171 … Z_emp1 0 161 128 : Z_emp4 0 1 050 431 Z_une1 0 17 846 : COMPOSITE Z_une4 0 67 283 CONTROL Z_nace1 0 115 624 TOTALS : Z_nace7 0 804 126 Z_nuts1 0 1 338 271 : Z_nuts8 0 22 555 
  • 15. May 15 - 16, 2014 15LFS Workshop in Rome Composite auxiliary variables  Overlapping part of the sample  Variables are taken from the previous wave of interview  Non-overlapping part of the sample  The values of variables are imputed
  • 16. May 15 - 16, 2014 16LFS Workshop in Rome Example 1. Overlapping January 2014 Wave 1 Wave 2 Wave 3 Wave 4 Wave 5 Previous interview none October 2013 October 2013 (July 2013) October 2013 Dependence: Theoretical overlap wave-to-wave is 4/5 (80%)
  • 17. May 15 - 16, 2014 17LFS Workshop in Rome Empirical results  We have compared the RC estimator to the GREG estimator in the FI-LFS real data (2006-2010)  Here we have used the ETOS program for point and variance estimation (Taylor linearisation method).  Relative efficiency (RE) can be formulated as  A value of RE greater than 100 indicates that the RC estimator is more efficient than the GREG estimator.    yrc ygr tV tV RE ˆˆ 100ˆˆ  
  • 18. May 15 - 16, 2014 18LFS Workshop in Rome Table 2. Distribution of calibrated weights for GREG and RC estimators (e.g 2nd quarter of 2006) The calibrated weights are obtained by the ETOS program. The results show that the variation of the RC weights is smaller than that of the GREG weights. GREG RCStatistics for calibrated weights Minimum 42.81 50.66 Maximum 416.29 221.87 Average 137.69 137.69 Median 135.12 137.15
  • 19. May 15 - 16, 2014 19LFS Workshop in Rome Table 3. Relative efficiency (RE, %) of estimates for the quarterly level of employment and unemployment by sex Quarterly level estimates Labour force status Sex RE (%) Employed Male 184,8 Female 178,6 Both sexes 173,5 Unemployed Male 114,2 Female 110,3 Both sexes 106,4
  • 20. May 15 - 16, 2014 20LFS Workshop in Rome Table 4. Relative efficiency (RE, %) of estimates for the monthly level of employment and unemployment by industrial classification Monthly level estimates NACE Sample size RE (%) Agriculture 251 395,6 Manufacturing 1 075 461,4 Construction 375 373,6 Wholesale and retail trade 899 318,3 Transport, storage and communication 390 365,4 Financial intermediation 802 398,3 Public administration 1 865 361,1
  • 21. May 15 - 16, 2014 21LFS Workshop in Rome Table 5. Relative efficiency (RE, %) of estimates for the quarterly level of employment and unemployment by industrial classification Quarterly level estimates NACE Sample size RE (%) Agriculture 788 358,2 Manufacturing 3 287 406,6 Construction 1 098 375,7 Wholesale and retail trade 2 698 375,6 Transport, storage and communication 1 216 369,1 Financial intermediation 2 403 370,8 Public administration 5 951 361,3
  • 22. May 15 - 16, 2014 22LFS Workshop in Rome Conclusions (1)  For the variables that were included as composite control totals, there are substantial gains in efficiency for estimates  For some variables it is future possible to publish monthly estimates where only quarterly estimates are published now?  Leading to internal consistency of estimates  Employment + Unemployment = Labour Force  Labour Force + Not In Labour Force = Population 15 to 74
  • 23. May 15 - 16, 2014 23LFS Workshop in Rome Conclusions (2)  It can be performed by using, with minor modification, standard software for GREG estimation, such as ETOS  It yields a single set of estimation weights  The results are well comparable with results reported from other countries  Chen and Liu (2002): the Canadian LFS  Bell (2001): the Australian LFS
  • 24. May 15 - 16, 2014 24LFS Workshop in Rome Future work  Analysis of potential imputation methods for the non- overlapping part of the sample?  Analysis of alternative variance estimators (Dever and Valliant, 2010)?  Incorporating information from all potential previous waves of interview
  • 25. May 15 - 16, 2014 25LFS Workshop in Rome MAIN REFERENCES BEAUMONT, J.-F. and BOCCI, C. (2005). A Refinement of the Regression Composite Estimator in the Labour Force Survey for Change Estimates. SSC Annual Meeting, Proceedings of the Survey Methods Section, June 2005. CHEN, E.J. and LIU, T.P. (2002). Choices of Alpha Value in Regression Composite Estimation for the Canadian Labour Force Survey: Impacts and Evaluation. Methodology Branch Working Paper, HSMD-2002-005E, Statistics Canada. DEVER, A.D., and VALLIANT, R. (2010). A Comparison of Variance Estimators for Poststratification to Estimated Control Totals. Survey Methodology, 36, 45-56. FULLER, W.A., and RAO, J.N.K. (2001). A Regression Composite Estimator with Application to the Canadian Labour Force Survey. Survey Methodology, 27, 45-51. GAMBINO, J., KENNEDY, B., and SINGH, M.P. (2001). Regression Composite Estimation for the Canadian Labour Force Survey: Evaluation ja Implementation. Survey Methodology, 27, 65-74. SALONEN, R. (2007). Regression Composite Estimation with Application to the Finnish Labour Force Survey. Statistics in Transition, 8, 503-517. SINGH, A.C., KENNEDY, B., and WU, S. (2001). Regression Composite Estimation for the Canadian Labour Force Survey with a Rotating Panel Design. Survey Methodology, 27, 33-44.