Statistical Journal of the IAOS

This paper argues that the Global Indicator Framework required to support the 2030 Agenda Sustainable Development Goals will not be successfully populated, using only existing approaches and mechanisms. Official statistical systems must adapt and consider new approaches if only partial success is to be averted. This paper presents a proposal to accredit unofficial statistics as official for the purposes of compiling sustainable development goal indicators. While there may be some reluctance, and there are certainly risks with this proposal, the arguments put forward highlight the potential for collaboration.

This paper looks at some of the ethical challenges that Big Data and data linking present for official statistics, reviews some of the international initiatives in train, focusing on recent developments in the UK. It emphasises the importance of confidentiality in maintaining trust in official statistics, arguing that, although the ethical issues are everywhere acknowledged as important, the debate on this is still at an early stage, and needs to be pursued.

Almost every public sector department produces some statistics and accumulates its share in the formulation of National Statistics. The accurate and timely statistics are vital for planning and development, budgeting and evaluation of the implemented programs. It may be reasonable to assume that datasets are being produced at almost all levels of the departments; nonetheless, a substantial number of valuable data-items are left unreported and hence they are unable to play their role in evidence-based planning and decision-making. There is a need to uncover these sources, to explore the reasons behind the non-reporting of data and to devise strategies to utilize these sources in the production of official statistics. In this paper, we present the results of a national-level survey conducted to collect information on data-processing and reporting mechanisms of public-sector organizations in Pakistan. Along with presenting the survey results, the paper discusses the potential sources of unreported data (including Big Data), the reasons for non-reporting at different sectoral levels and the confidentiality, privacy, and data-sharing constraints. Based on the above, the paper ends by proposing the compilation of a Directory of Administrative Data Sources (DADS) in order to establish an improved administrative infrastructure in the country.

Since 2009 Statistics Netherlands (SN) has been exploring the use of Big Data (BD) in official statistics. This research resulted in new or improved official statistical products such as the Consumer Price Index based on web scraped prices, and the traffic intensity indicator using traffic loop data. With the aim to make use of new BD sources, SN established the Center for Big Data Statistics (CBDS). The CBDS builds upon acquired data science expertise in an ecosystem of academic and private partners in order to make official statistics using survey, administrative and BD sources and methods. This multisource approach provides better insights into complex policy questions. One of these complex questions for policy makers is the energy transition and how to move away from carbon based to renewable energy sources. More detailed data is necessary in order to get a better insight into locally produced renewable energy and to allow policy makers to design policies that allow matching offer and demand. We use multiple data sources in a combined fashion to provide insight into solar energy provision. We conclude by giving general guidelines for the process to embed the experimental statistic in the official statistical process.

This paper discusses the opportunities and challenges presented when utilizing scanner data to compile the CPI, with a focus on implementation guidance to national statistical offices. Empirical results for the Australian CPI are examined following the implementation of a multilateral price index method in December quarter 2017.

According to monetarists, inflation is always and everywhere a monetary phenomenon. It may, however, have supply side drivers; including global commodity prices. In order to have better understanding of inflation dynamics in a small open economy like Pakistan, we need some new sub-indices within CPI basket. In this paper, we have proposed and estimated new subgroup indices within CPI basket of Pakistan for monthly data from July 1991 to June 2018. These sub-indices include those for ‘services’ and ‘goods’ prices. Within goods group, we have further bifurcated the sub-indices for ‘tradable’ and ‘non-tradable’ goods prices. Tradable goods price index is further bifurcated into ‘food and energy’ and ‘non-food and non-energy’ tradable sub groups. On average, goods account for two thirds of our CPI basket and remaining one third belongs to services. Non-tradable goods and services comprises of three-fourths of overall consumer basket, while one-fourth are tradable goods. Share of tradable goods in CPI basket has increased over time. Inflation rate in tradable goods prices is volatile and lacks persistence whereas services prices are found relatively stable and highly persistent in Pakistan. Inflation in tradable goods basket is found to lead inflation in non-tradable prices.

The paper presents a series of reflections deriving from teaching official statistics. Much of the accumulated experience derives from teaching in the “Methods and tools for official statistics” and “Survey methods: traditional and new techniques in Official Statistics” courses in the European Master in Official Statistics (EMOS) at the universities of Florence and Pisa, as well as on numerous occasions at tertiary education centres. Such experiences highlighted the lack of the themes in question within the standard study plans and the scarce awareness of students about the official statistics; and allowed the identification of the essential topics to be addressed. Four basic pillars (definitions, quality criteria and practices, sources, and autonomy in finding official data) need to be spread much more than only by statistical authorities. Manuals on basic statistics (data-science) should always foresee one or more chapters dedicated to the recognition of quality data and official statistics. Teachers from any discipline which implies the transmission of the value of high quality statistical information need to be trained on this aspect. The general insight of the experience of teaching official statistics is that if data science may remain a specialist knowledge, statistical literacy needs to become a common one.

Data integration is becoming a crucial task in National Statistical Institutes in order to exploit the information provided by already existing data sources. Here the focus is on statistical matching methods; they are designed to integrate data stemming out from traditional sample surveys referred to the same target population. In particular, this work shows how popular statistical learning techniques can be beneficial for matching purposes. Two proposals are presented, having a different final scope: the creation of a “fused” data set or the assessment of the uncertainty due to the typical statistical matching scenario. The characteristics of these procedures are investigated through a series of simulations and in an application to real survey data. The achieved results are encouraging and show that some statistical learning techniques can be very effective in exploiting the information provided by already existing survey data, permitting a reduction of the uncertainty determined by the typical statistical matching setting.

Data from the National Health and Nutrition Examination Survey (NHANES) have been linked to the Center for Medicare and Medicaid Services’ Medicaid Enrollment and Claims Files. As not all survey participants provide sufficient information to be eligible for record linkage, linked data often includes fewer records than the original survey data. This project presents an application of multiple imputation (MI) for handling missing Medicaid/CHIP status due to linkage refusals in linked NHANES-Medicaid data using the linked 1999–2004 NHANES data. By examining multiple outcomes and subgroups among children, the analyses compare the results from a multi-purpose dataset produced from a single MI model to that of individualized MI models. Outcomes examined here include obesity, untreated dental caries, attention deficit hyperactivity disorder (ADHD), and exposure to second hand smoke.

Almost 2 years have passed since the launching of the 2030 Sustainable Development Goals but no big strides have been made especially on data compilation to track and measure these goals. Uganda, like the rest of the world has to base its development agenda on SDGs. However, the major constraint often lies on measuring the indicators and even compilation of data. The author discusses 7 SDG economic indicators based on the Ugandan context. This paper considered a baseline year of 2014 for the economic indicators of the SDGs in Uganda. A comparative review of data was obtained from UNSD SDG indicator global database and available national data in Uganda to update the Uganda country page constructed by the ACS survey in 2017. Metadata analysis was invoked following a 4-assessment check. Results showed that a lot of data from administrative sources could be used to feed into the Ugandan page given that censuses and surveys are not regularly updated. However, the paper shows some discrepancies in the ACS survey report data. There is need to explore more of the administrative data in government ministries and agencies in order to close the data gaps for the economic indicators of the SDGs.

The article presents an international comparison of selected quality of employment indicators from the perspective of the employed person. The conceptual framework is based on the Handbook on Measuring Quality of Employment, A Statistical Framework, released by the United Nations Economic Commission for Europe (UNECE). From the 67 indicators presented by the framework, 21 were selected for the comparison between Finland, Germany, Israel, Italy and Switzerland, ensuring that all dimensions for measuring quality of employment were covered and analyzing sex and age differences for several indicators. The present article shows the main results for the indicators chosen and the limits of the international comparability, mainly related to different institutional and labour market contexts, and to the heterogeneity of data sources.

Dual-system estimation is a well-established approach for estimating an unknown population size from two independent but imperfect counts of the population. In this paper we develop the estimation framework for using a coverage survey and population census as the two sources and combining with ratio estimation to produce a set of population estimates. Adjustments are developed to correct for a failure of the key assumptions of homogeneity and independence that under-pin dual-system estimation using an external count of the number of households. The issue of over-count within the census is also discussed and a bootstrap approach to variance estimation is proposed. A comprehensive set of simulation results are presented to support the decision to implement the framework to estimate the population following the 2011 Census of England and Wales; and the implementation to the estimation of census coverage in 2011 is discussed.

Study of financial bubbles is the extremely important topic in the modern society. Their formation and dramatic bursts are frequently considered to have a massive impact on most of the fields all over the world. Although the literature presents plenty of reviews on bubbles, crashes, and financial crises, the debate is still open even on whether or not bubbles can persist in modern asset markets. The idea of usage of econometric tests to detect these bubbles are not new and can be classified into 6 groups namely, tests based on the variance, tests based on unit root, tests based on regimes, tests based on Johansen-Ledoit-Sornette model, tests based on durations and tests based on neural networks. This paper presents a review of research in these areas of detection and analysis of the financial bubbles.