Estimates for geographical domains through geoadditive models in presence of missing geographical information
Chiara Bocci, Emilia Rocco
This paper deals with the matter of applying a geoadditive model to produce estimates for some geographical domains in the absence of point referenced geographical data. The implementation of a geoadditive model needs the statistical units to be referenced at point locations and if we use it to produce model-based estimates of a parameter of interest for some geographical domains, the spatial location is required for all the population units. This information is not always easily available. Typically, we know the coordinates for sampled units, but for the non-sampled units we only know the areas - like blocks, municipalities, etc. - to which they belong. In such situation, the classic approach is to locate all the non-sampled units by the coordinates of their corresponding area centroid. This is obviously an approximation and its effect on the estimates can be strong, depending on the level of nonlinearity in the spatial pattern and on the area dimension. We propose a different approach that, instead of using the same coordinates for all the units, imposes a distribution for the locations inside each area. Our approach is formalized under a Bayes inferential perspective and its performance is evaluated through various Markov Chain Monte Carlo experiments implemented under different scenarios.
Published as Statistical Methods & Applications (Online First Articles), pp. 1-23, 2014; link, published.
wp2011_02 (Statistics for experimental and technological research)
The assessment of mixed and intermediate foci images using the R software environment
Claudio Procaccianti, Chiara Urani, Federico M. Stefanini
Fast and reliable screening of the carcinogenic potential of a chemical compound can be performed using in vitro methods such as the cell transformation assay (CTA), in which selected cell lines grow under different treatment conditions (with/without chemical under evaluation) and colonies (foci) formed at the end of the assay are scored and classified by light microscopy. While foci can mostly be divided into three canonic classes (Type I, II and III), often more undefined phenotypes can be spotted, resulting in an uncertain class attribution (mixed and intermediate). Here, we describe the R code developed to calculate a quantitative dissimilarity index and classify mixed or intermediate foci by exploiting the quantitative information provided by digital images of foci colonies.
Multiplicative Error Models
Christian T. Brownlees, Fabrizio Cipollini, Giampiero M. Gallo
Financial time series analysis has focused on data related to market trading activity. Next to the modeling of the conditional variance of returns within the GARCH family of models, recent attention has been devoted to other variables: first, and foremost, volatility measured on the basis of ultra-high frequency data, but also volumes, number of trades, durations. In this paper, we examine a class of models, named Multiplicative Error Models, which are particularly suited to model such non-negative time series. We discuss the univariate specification, by considering the base choices for the conditional expectation and the error term. We provide also a general framework, allowing for richer specifications of the conditional mean. The outcome is a novel MEM (called Composite MEM) which is reminiscent of the short- and long-run component GARCH model by Engle and Lee (1999). Inference issues are discussed relative to Maximum Likelihood and Generalized Method of Moments estimation. In the application, we show the regularity in parameter estimates and forecasting performance obtainable by applying the MEM to the realized kernel volatility of components of the S&P100 index. We suggest extensions of the base model by enlarging the information set and adopting a multivariate specification.
Published as Ch. 9 in Bawens L., C. Hafner and S. Laurent, Handbook of Volatility Models and Their Applications, J. Wiley, pp. -, 2012, ISBN 978-047087251.
A home to plan the first child? Fertility intentions and housing conditions in Italy
Daniele Vignoli, Francesca Rinesi, Eleonora Mussino
In Italy, the existing literature on the link between low fertility in Italy and people’s socio-economic status has been surprisingly silent about the potential role of housing. This study aims at filling this gap: Are Italians who are better off in terms of housing security more likely to plan to have a first child in the short run? Our results suggest that there is a clear positive gradient between the fertility intentions of couples and the degree to which they feel secure about their housing situation. The positive gradient remained substantial even after we controlled for the demographic and socio-economic characteristics of the couple, and after we adjusted the estimates for the presence of unobserved characteristics and the endogeneity between housing and fertility. This implies that, in Italy, the housing dimension should not be disregarded when studying fertility differentials.
Published as Population, Space and Place, Volume 19, pp. 60-71, 2013; link, published.
Identification of causal effects in the presence of nonignorable missing outcome values
Alessandra Mattei, Fabrizia Mealli, Barbara Pacini
We consider a new approach to identify the causal effects of a binary treatment when the outcome is missing on a subset of units and dependence of nonresponse on the outcome cannot be ruled out even after conditioning on observed covariates. We provide sufficient conditions under which the availability of a binary instrument for nonresponse allows us to partially identify causal effects in some latent subgroups of units, defined by their nonresponse behavior in all possible combinations of treatment and instrument, named Principal Strata. Examples are provided as possible scenarios where our assumptions may be plausible; they are used to discuss the key role of the instrument for nonresponse in identifying average causal effects in presence of nonignorable missing outcomes and provide new guidelines on study designs for causal inference.
Human Capital and the Entry to Motherhood in Italy
Increasingly, young women educate themselves for a lifelong professional career. The goal of this study is to add to our understanding of differences in the timing of first births by women’s human capital. We focus on Italy, a country where the process of fertility aging is particularly advanced. Applying intensity regressions to the Italian Household Multipurpose Survey of Family and Social Subjects our results show that paid employment strongly conflicts with motherhood in the country. Marked educational differentials in the way women’s employment affects transition to motherhood were found, however. The conflict seems much stronger for medium- and low- educated women, while the tertiary educated ones are more likely to conceive their first child if they are employed. Finally, this research provides evidence that conditions for work and family reconciliation, although important, are not the only factors leading to fertility postponement.
Gender Equality and Fertility: Which Equality Matters?
Gerda Neyer, Trude Lappegård, Daniele Vignoli
Does gender equality matter for fertility? Demographic findings to this question are rather inconclusive. We argue that gender equality is a complex issue that needs to be conceptualized in a way which includes gender equity and allows for gender differences but uncovers gender inequalities. We explore this approach by investigating the impact of four dimensions of gender equality on women´s and men´s childbearing intentions in Europe: the possibility to maintain a household, the capabilities to choose, the resources to have agency, and gender equity in household work and in care. We apply logistic regressions to data of the Generations and Gender Survey. Our results suggest that gender equality and fertility intentions are intertwined in a multi-faceted way, and that gender equality in the areas which we examine exert different impacts on women’s and men’s childbearing intentions. Our study also confirms that parenthood still constitutes a dividing line between more and less gender equality, and that this affects childbearing intentions of childless women and childless men differently than that of mothers and fathers. This necessitates an approach which allows identifying the essential gender inequalities in employment, in society, and in the family which matter for childbearing decisions.
Whose economic instability affects the likelihood of becoming a parent in Italy? A tale of two partners
Daniele Vignoli, Sven Drefahl, Gustavo De Santis
We examine fertility differentials in Italy taking into account income and employment (in)security of both partners in a couple. We use data from four waves of the Italian section of the EU-SILC (Statistics on Income and Living Condition), 2004-2007, accounting for its longitudinal nature. Overall, our results suggest that Italian couples are neither fully traditional nor entirely modern: the "first pillar" (i.e. a male partner with a job) is still crucial in directing fertility decisions, probably because it gives the household a feeling of (relative) economic security. Once that is granted, the second most important variable is not income: rather, it is the position of both partners in the labour market, and how secure their own job is. Indeed, when both partners have jobs of unlimited duration, fertility is highest; with alternative combinations, fertility is lowered considerably.
Published as Demographic Research, Volume 26, Issue 1, pp. 41-62, 2012; link, published.
Different women’s employment and fertility behaviours in similar institutional settings: Evidence from Italy and Poland
Anna Matysiak, Daniele Vignoli
In this paper we compare Italy and Poland, two countries where the country-specific obstacles to work and family reconciliation are similarly strong, but which differ in terms of the history of women’s labour force participation and of household living standards. We adopt a life course perspective, and trace women’s employment choices around the first and the second birth. On the one hand, our findings suggest the presence of a strong conflict between women’s paid work and childbearing in both countries. On the other, our results show that women’s employment clearly inhibits childbearing in Italy, while in Poland women tend to combine the two activities. Overall, we find that countries characterised by similarly strong institutionally or culturally driven tensions between work and family may differ in how women’s fertility and employment behaviours are interrelated.
wp2011_10 (Economic Statistics)
Productivity change of Italian firms: an analysis of panel data using log linear models
Laura Grassini, Gianni Marliani
A provisional version of panel data, built in collaboration with Istat (the Italian National Statistical Institute), is used to synthetically describe the productivity profile of Italian manufacturing firms in different economic sectors (Ateco divisions). For each sector, a mobility table is built by classifying firms according to the categorized values of a labor-productivity index in two different years (1998 and 2004). The analysis is carried out using log-linear models, in order to identify specific association patterns (inertia, downgrading, upgrading) within these mobility tables. The empirical analysis reveals typical patterns in some economic sectors but, on the whole, it shows an objective difficulty in interpreting the results.
Statistical Methods for Understanding Hydrologic Change
Chiara Bocci, Enrica Caporali, Alessandra Petrucci
Rainfall is considered a highly valuable climatological resources and is a fundamental component of any water resource assessment strategy. Rainfall characterization has important repercussions, among others, on climate change scenarios, flood risk mitigation, water resources availability and drought assessment. We use a hierarchical modeling approach to investigate a collection of spatially referenced time series of extreme values. We implement a geoadditive mixed model for extremes with a temporal random effect, where we assume that the observations follow generalized extreme value (GEV) distributions whose locations are spatially dependent where the dependence is captured using the geoadditive model. The analyzed territory is the catchment area of Arno River in Tuscany in Central Italy. The dataset is composed by the time series of annual maxima of daily rainfall recorded in about 400 rain gauges, spatially distributed over an area of about 8.830 km2 . The record period covers mainly the second half of 20th century.
Published as Advances in Statistical Analysis, 2012; link, forthcoming.
wp2011_12 (Statistics for experimental and technological research, Statistics)
Analisi delle scelte modali per la mobilità fiorentina: un esperimento di scelta
Rossella Berni, Fabrizia Mealli
In questo lavoro si presentano i risultati di un esperimento di scelta volto a valutare le preferenze degli utenti, che si spostano nell'area fiorentina per motivi prevalentemente di studio e lavoro, rispetto ad alcuni mezzi di trasporto. Dopo una prima rassegna sugli esperimenti di scelta e sui modelli multinomiali per scelte discrete, il caso di studio è presentato nel dettaglio considerando: il disegno sperimentale proposto, l'esperimento di scelta somministrato, i risultati delle stime dei modelli a scelta discreta ottenuti sui dati raccolti. I risultati evidenziano alcune rigidità nelle preferenze, e dunque la difficoltà a spostare gli utenti da alcune tipologie di mezzi di trasporto ad altre, agendo sulle componenti relative al costo del mezzo ed ai tempi di spostamento.
The educational gradient in marital disruption: A meta-analysis of European longitudinal research
Anna Matysiak, Marta Styrc, Daniele Vignoli
A large number of empirical studies have investigated the role of education in the changes in union dissolution in Europe, but these studies have so far produced inconsistent results. This paper seeks to assess the relationship between educational attainment and the incidence of marital dissolution by systematizing the existing empirical evidence on the topic. To this end, we have conducted a quantitative literature review (a meta-analysis). This review allowed us to assess the relationship in a quantitative manner and to investigate its temporal change, net of the across-study differences. Our results illustrated that a reversal in the educational gradient from positive to negative has occurred over time. The findings also showed that the change in the educational gradient was happening in parallel to an increase in access to divorce. Finally, the findings suggested that women’s empowerment has played a greater role in explaining the changing educational gradient of divorce than the liberalization of divorce laws.
wp2011_14 (Statistics, Statistics for experimental and technological research)
Evaluation of Kinship Identification Systems Based on STR DNA Profiles
Fabio Corradi, Federico Ricciardi
In this paper we detail how to evaluate a kinship identification system, a probabilistic tool devoted to obtain the Likelihood Ratio required in deciding if a candidate is a specific member of a family given some genetic profiles observed in the familial pedigree. The paper considers the LR as a random variable, depending on the still unobserved genetic DNA evidence of a candidate to identification, posing attention to the familial expected possibilities to identify a specific member. In a decision theory perspective, we evaluate which system, among a set of possible alternatives, is the most suitable to fulfil the requirements of the parts involved. The proposed system evaluation, carried on before the identification trial is performed, is specific for each case and does not require any additional laboratory expenses, since it makes use of a subset of the employed data. Special attention is devoted to the computational aspects of the implied high dimensional space problem: matters concerning approximations are discussed. A case study illustrates how the approach proves to be especially helpful when the distance in the pedigree between the observed DNA donors and the unobserved relative possibly identifying a candidate is large.
wp2011_15 (Demography, Statistics)
Fertility intentions and outcomes. Implementing the Theory of Planned Behavior with graphical models.
Letizia Mencarini, Daniele Vignoli, Anna Gottard
This paper studies fertility intentions and their outcomes. We derive the theoretical model from the Theory of Planned Behavior (TPB), and test its validity in a low-fertility context (i.e., Italy). We move beyond existing research and use the methodology of graphical models to generate a precise picture and a formal description of the developmental fertility decision-making process, representing the possible dependencies between all of the involved variables by a chain graph. Overall, this analysis strongly supports the validity of TPB, and therefore contributes to the systematic study of fertility decision-making processes. However, our findings also produce some evidence that challenges and complicates the application of the TPB to fertility research, especially regarding the role of background factors.
Ultimo aggiornamento 26 febbraio 2014.