Poster Session 1

Back to Schedule


Presenting AuthorAbstract
Mostafa Younes
e-mail: mostafa_y@capmas.gov.eg
Title: <<< Procedures to ensure the quality of HICS survey data in Egypt >>>
The concept of data quality covers several aspects, from the planning of the survey to the end of it, dissemination and utilization of the data. The most important components of statistical quality elements are: accuracy, comparability, and quality control procedures. Data accuracy includes multiple aspects of any survey, could divided to sampling errors and non-sampling errors. Non-sampling errors returns to the way the work is performed and the survey tools. The national statistical office of Egypt (CAPMAS) conducting the household income, expenditure and consumption survey (HICS) survey every 2 years Started from 2008, which aim to better monitoring the changes in standard of living of the family. The present survey 2017 is Thirteenth in the series of these HICS surveys. Wherefore, there are a number of actions taken to reduce non-sampling errors in the household income, expenditure and consumption survey (HICS).The paper will review the Quality control procedures carried out on the survey, starting from the selection of the fieldworkers, field work training, income and expenditure unit duties, call back unit, the field work quality team and field work technical team. The household income, expenditure and consumption survey (HICS) 2015 response rate was 95.9% when the sample was 25000 households and was designed to be representative at the governorate level, have been visited for two research periods.
Eleni Ketzaki
e-mail: eketzaki@yahoo.gr
Title: <<< A method on reducing small sample bias of the obtained as a matrix product Gini Index. >>>
Gini Index is one of the most commonly used measure of income inequality. It has been noticed that it is a downward biased measure in small populations. In this study we propose an expression that aims to reduce bias, in case that the Gini Index is interpreted as a matrix product. First of all we examine the bias correction for a number of individuals and then we investigate the bias due to grouping. The formulation of the Gini Index as a matrix product has been chosen because it allows easier and faster computation. Moreover the calculation of the Gini Index based on data grouped by categories, commonly arises with income data, that are usually grouped for confidentiality purposes. In the end of our study we provide a numerical example in order to compare the proposed method with the one that already exists regarding to the bias correction.
Álvaro Gómez Losada
e-mail: alvaro.gomez-losada@ec.europa.eu
Title: <<< A novel approach to detecting pattern changes in time series from Eurostat >>>
Detecting pattern changes in time series (TS) is of the utmost importance when evaluating a phenomenon’s trend shifts over time. In particular, studying these changes on the latest observations of any TS can help to inform decisions as to what actions to take on a particular issue. The methodology for formalizing detection of these changes at the time an official estimate is released for a TS should be based on well-settled, appropriate methodology. This study introduces a new method based on Hidden Markov models (HMM) to ascertain whether the latest figures of a TS might be experiencing a shift or structural break with respect to the regular trend. Ultimately, these latest TS observations are the most important for policy makers. HMM were used to detect regimes (clusters) in TS and to characterize them using their average and standard deviation values. These latest observations of TS are considered not to experience a shift when their values are within a confidence interval (CI) estimated using those average and standard deviation values from their regimes. This CI can be modified to detect when the last observations are experiencing a shift or structural break with regard to the TS trend. This methodology is modular and flexible, allowing any TS to be studied according to their characteristics over time. The R software was used to deploy this methodology, and TS with different lengths and characteristics from Eurostat were tested. Results obtained show promising sensitivity results in detecting shifts or structural breaks in TS.
Natasha Cenova
e-mail: natasa.cenova@stat.gov.mk
Title: <<< Improving the quality and relevance of the official statistics produced in the SSO of the Republic of Macedonia >>>
This paper, in the context of improving the quality of collected data on issued building permits, analyses the quality and relevance of the official statistics. Data on issued building permits are collected according to the Eurostat Methodology of short-term business statistics, Council Regulation (EC) 1165/98, Commission Regulation (EC) 1503/2006. The indicators for issued building permits are part of the PEEI main economic indicators for the economic developments in the EU. They also provide important information on the economy and monetary policy as well as for future investments in the construction sector. In the Republic of Macedonia, the survey was implemented in 2008 with paper questionnaires sent to the authorities responsible for building permits. In 2014, the Government of the Republic of Macedonia started a project for online data collection via an electronic system for issuing building permits, e-permits, prepared by the responsible authorities - the Association of the Units of Local Self-Government (ZELS) together with the Ministry of Transport and Communication. In cooperation with these institutions, the State Statistical Office incorporated the variables needed for the survey by adding a questionnaire for the official statistics. The reporting units provide information on the required variables directly on the e-permits website while applying for a permit. In this manner, the quality and coverage of the collected data from administrative registers is improved and the relevance of the official statistics is raised to a higher level. The published data on issued building permits are used by the Government, the business community, international organisations, etc.
Andrés García Carreira
e-mail: andres.garcia.carreira@ine.es
Title: <<< The ESA 2010 Pensions Table: an integrated view on the functioning of pension systems in Spain >>>
The inexorable impact of the population ageing, the peculiarities of pay-as-you-go schemes of public systems and the increasing role played by private systems in developed societies emphasize the need of a harmonized measure of right and obligations accrue in them as one the main priorities for the statistical systems. Current national accounts standards (SNA 2008 and ESA 2010) already include guidelines for the registration in their systems of all private pension obligations/rights linked to employment regardless of whether they are systems with or without constitution of reserves. In addition, they propose the recording of all pension schemes, including contingent obligations/rights accrued in public systems in an annexed table. The supplementary table for pension rights accrued to date in social insurance schemes will allow us to see the evolution of all pension rights stocks and the flows that motivate their variations, regardless the fact they are non contingent financial assets/liabilities for the households/pension managers or not. Both the objectives and data compiled in the table present obvious conceptual difficulties and require a high level of expert knowledge in the financial, insurance and actuarial fields. Thus, in the Spanish case, the close collaboration with external agencies from various areas has been a basic component of the project, as a clear example of interinstitutional cooperation towards the highest standards of quality in official statistics. Furthermore, a didactic dissemination of the pension tables results as a tool for analyzing the functioning of national pensions systems but not as a measure of their future feasibility is one of the most challenging issues that ESS and other international organizations face nowadays.

Back to Schedule

Font Resize
Contrast