Marco Fortini

Italian National Institute of Statistics · Censuses, Adminstrative and Statistical Registe

Record linkage addresses the problem of identifying pairs of records coming from different sources and referred to the same unit of interest. Fellegi and Sunter propose an optimal statistical test in order to assign the match status to the candidate pairs, in which the needed parameters are obtained through EM algorithm directly applied to the set...

RELAIS (REcord Linkage At IStat) is a toolkit providing a set of techniques for dealing with record linkage projects. The principal features of RELAIS are: a) the combination of different techniques for each of the record linkage phases, so that the resulting work-flow is actually built on the basis of application and data specific requirements; b)...

In the last decades Italy has become an immigration country. The importance of studying this phenomenon and the characteristics of the population involved are pressuring the official statistics on the need of reliable data. In 2011 thanks to the innovation of the 15th population and housing census and the use of the auxiliary list (LIFA) we were al...

The paper assesses the performance of administrative archives in reducing under-coverage errors in register-assisted population censuses caused by poorly maintained population registers. Preliminary results from a simulation based on the 2009 pilot survey provide encouraging evidence on the usefulness of these auxiliary archives.

In this paper we present a mixture model for the identification of cheating interviewers in the social surveys carried out by Istat. An EM algorithm is used to estimate the parameters of the model, starting from reinterviews results and from a set of variables describing the interviewer characteristics and performance. The above method is applied t...

The combined use of statistical survey and administ rative data is largely widespread to maximize their respective usefulness: unfortunately data sources are often hard to integrate due to errors or lacking informat ion. Record linkage techniques are a multidisciplinary set of methods and practices aimi ng to identify the same real world entity, di...

Riassunto. Una qualsiasi indagine statistica è soggetta a mancate risposte, errori di copertura e di misurazione. Nel caso di indagini di elevato interesse pubblico viene spesso condotta una seconda rilevazione con il duplice obiettivo di valutare l'entità dell'informazione mancante e la precisione delle misurazioni. In questi casi, per la validazi...

Record linkage has been subject of research for several decades, and a huge number of record linkage solutions have been proposed, based on probabilistic and empirical paradigms. However, record linkage is a complex process, for the execution of which one single technique is often not enough; it can be seen as composed by distinct phases, each requ...

In this paper is shown a Bayesian approach to the capture recapture estimation of a population amount when the data are affected by record linkage errors. Though the method is of general employment, it is considered here mainly with reference to post enumeration surveys for the population census coverage evaluation. The method considers the case of...

Record Linkage (RL) refers to the use of an algorith- mic technique to match records from dierent data sets that correspond to the same statistical unit. RL is ubiquitous in ocial statistics: estimation of pop- ulation size via capture-recapture methods, testing of disclosure strategies and coverage measurement surveys are only few examples. A key...

This paper provides an overview of the latent class models and of the Bayesian method to estimate response error and reports an application to real data. The results are then compared with those obtained by the classical reinterview methodology. Our paper suggests that, in presence of accurate prior information about the error, the Bayesian approac...

Information about the presence of non-sampling errors affecting survey data is of importance both for final users and for survey managers. Besides the direct methods for evaluating the effects of non-sampling errors on survey estimates, the analysis of data obtained from each phase of the production process represents an easy and inexpensive way to...

To identify disease patterns in AIDS-related focal brain lesions (FBL) and to design a decision-making strategy for differential diagnosis. Prospective study. Probabilities of CNS disorders were calculated using Bayes' theorem according to clinical variables (mass effect at CT or MRI, Toxoplasma serology, anti-Toxoplasma prophylaxis) and to the res...

Anticholinergic anti-Parkinson drugs are often used to prevent or treat neuroleptic-induced extrapyramidal side effects. National Health Service prescriptions for these drugs and haloperidol delivered in the geographical area of Rome, Italy, and the surrounding province (about 3,750,000 inhabitants) from 1986 to 1989 were reviewed. Individual presc...

The epidemiological use of drug prescription data is getting increasing attention in Italy. All Italian citizens are included in the National Health Service (NHS) which reimburses almost every drug prescription. The potential information content of such systems have to be evaluated within a technological context which will allow, for medium term, d...

This paper proposes a method of moments for estimating simple and correlated measurement variance componentswhen a reinterview is available on a subsample of respondents, but the two measurements cannot be considered as being collected under the same conditions and they are therefore subject to different measurement error variance.This seems a more...

The combined use of statistical survey and administrative data is largely widespread to maximize their respective usefulness: unfortunately data sources are often hard to integrate due to errors or lacking information. Record linkage techniques are a multidisciplinary set of methods and practices aiming to identify the same real world entity, diffe...