We introduce a distribution-free approach to quantile share ratio regression. Our proposal involves the specification of a generalised linear model for the ratio of tail areas above and below two pre-specified quantiles. The latter ratio is the quantile share ratio, a measure of primary interest in the study of income inequality. We derive inference through an efficient two-step approach for parameter estimation that entails estimation of the conditional cumulative distribution function at the first step. A scalable strategy is discussed for large sample sizes. We are motivated by the study of income inequality in the European Union. Using data from a sample of approximately 2.8 million households across twenty-three countries and fifteen years (2007-2021) we make formal claims on the significance of adjusted and unadjusted differences among countries, and time trends. Interestingly enough, we find independent negative associations of economic inequality with gender equality and control of corruption.

Direct oral anticoagulants (DOACs) reduce thromboembolism in atrial fibrillation (AF), but their effect on cardiac outcomes is less studied. Systematic review and network meta-analysis were performed on AF patients on DOACs/vitamin K antagonists (VKAs). Both observational studies and randomized clinical trials (RCTs) were included. Endpoints were myocardial infarction (MI) and major adverse cardiac events (MACE). Rankograms and SUCRA were performed. Sub-group analysis included age (

This paper examines temporary migration and return decisions among immigrants in Italy using a novel administrative dataset covering 3.7 million foreign-born individuals between 2011 and 2022. By reconstructing individual migration histories, we estimate migration duration using parametric survival models, quantile regressions for interval-censored data, competing risk models, and a split cure model that distinguishes permanent settlement from the timing of exit. Results show that out-migration is concentrated in the first five years after arrival, while most migrants remain in Italy over the 12-year observation window. Age and gender matter, but local conditions within Italy strongly shape migration duration. Higher local incomes are associated with longer stays, while higher rental prices accelerate departures. Regional disparities also matter independently of economic variables: migrants in the South and Islands remain significantly longer than those in the North. These findings show that heterogeneity within host countries, rather than national averages alone, shapes migration trajectories and highlights the importance of local labor markets and living conditions.

We compared a novel deep learning (DL)-based quantification software (BTXBrain) with an established platform (Q.Brain) for quantitative perfusion single photon emission computed tomography (SPECT) in patients with clinically confirmed neuropsychiatric systemic lupus erythematosus (NPSLE). This retrospective exploratory inter-platform comparison study included patients with clinically established NPSLE who underwent brain SPECT with technetium-99m ethyl cysteinate dimer (99mTc-ECD). SPECT datasets were reprocessed using BTXBrain and Q.Brain. Regional perfusion estimates were compared across predefined cortical territories using paired analyses, variance ratio assessments, and Kendall’s tau correlation analysis. A total of 540 paired regional and hemispheric evaluations from 30 NPSLE patients were successfully processed and analyzed. BTXBrain and Q.Brain unveiled systematic, region-specific divergence in perfusion metrics. BTXBrain yielded higher standardized uptake value ratio values than Q.Brain in frontal (mean difference 0.081, p < 0.001), posterior cingulate (0.090, p < 0.001), medial temporal (0.333, p < 0.001), and occipital regions (0.078, p < 0.001). Conversely, a reversed trend was observed in the right lateral parietal cortex (-0.056, p = 0.009). Variance ratio analysis confirmed that the DL-based approach introduces a significantly different data dispersion profile (p < 0.05 across multiple macro-regions), while meaningful cross-platform correlation was restricted to specific territories, such as the left posterior cingulate cortex (τ = 0.44, p = 0.002). BTXBrain and Q.Brain are not directly interchangeable in patients with NPSLE. Consequently, rather than a generic drop-in replacement, the routine clinical adoption of DL-driven SPECT quantification requires software-specific reference thresholds and further calibration before cross-platform use.

We estimate the extent of conflict-related homicides, kidnappings, and forced recruitments in Colombia between 1985 and 2018; also at the gender, department, and year level. In order to do so, we first introduce general principles for stratified estimation. We then implement Ridge and fused Ridge type penalties to smooth generalized Chao and Zelterman estimators. The fused Ridge type penalties are particularly apt at pooling information across time and space, after specification of an opportune adjacency matrix. Penalty parameters are selected through an empirical Focused Information Criterion strategy. A simulation study illustrates how penalized estimators outperform competitors in terms of mean squared error, and stability, both overall and at the stratified level. We estimate more than 640,000 conflict-related homicides, almost 30,000 forced recruitments, and more than 60,000 conflict-related kidnappings in the period; with clear temporal and spatial patterns. The Antioquia department, and years between 2000 and 2005, were most severely affected by violence. We also estimate that after the peace agreements of 2016 in some departments the conflict experienced a slight resurgence in killings and, more broadly, in human rights violations.

We propose a robust feature-weighted jump model for time-dependent clustering. A penalty is used to encourage smoothness of transitions over time, while robustness is achieved through the use of a Tukey's biweight loss function. An additional parameter controls the variability of feature weights across states, allowing the model to assign state-specific relevance to each feature. We illustrate in simulation how the method accurately recovers the true cluster sequence and reliably identifies relevant features, outperforming competing approaches, particularly in the presence of outliers. We conclude with two empirical applications, one on the number of conflict-related homicides in Kosovo in the period 1998-2000, and another on macroeconomic performance of twelve European countries in the period 1949-2024.

We propose a regression model for cylindrical response variables that arise in the analysis of events occurring randomly over time. Each response consists of two components, one related to the timing of the event (circular), and one related to the intensity or the consequences of the event (linear). We use the multivariate generalized Laplace distribution whose parameters make this model more flexible than-as well as a generalization of-the ordinary multivariate Gaussian model. For inference, we propose standard maximum likelihood procedures. We investigate about 134,000 road traffic accidents from the Fatality Analysis Reporting System of the US National Highway Traffic Safety Administration. We define the circular component as the time of the day at which the accident occurs, and the linear component as the gravity of the event. The latter comprises an overall score for the severity of the injuries experienced by the people involved in the accidents, along with the median age of the victims. Our analysis shows that the timing and severity of the accidents are influenced by temporal and geographical factors.

We develop quantile ratio regression for panel data, to model covariate effects on ratios of upper and lower conditional quantiles. The proposed estimator is based on semi-parametric estimation of conditional quantiles, which are then linked to covariates via a linear model. Dependence within subjects is accommodated by a ridge-type penalty on unit-specific intercepts, which shrinks individual effects while allowing for heterogeneity. We also introduce a computationally efficient one-step empirical Bayes procedure for selecting the penalty parameter. A simulation study under different dependence structures and outcome distributions shows that the ridge-regularized estimator reduces mean squared error and improves out-of-sample prediction relative to unpenalized and naïve alternatives. An application to a four-year panel of about twenty thousand European households illustrates how the method can be used to analyze income inequality through conditional income quantile ratios.

We propose a MANOVA test for semicontinuous data that is applicable also when the dimension exceeds the sample size. The test statistic is obtained as a likelihood ratio, where the numerator and denominator are computed at the maxima of penalized likelihood functions under each hypothesis. Closed form solutions for the regularized estimators allow us to avoid computational overheads. We derive the null distribution using a permutation scheme. The power and level of the resulting test are evaluated in a simulation study. We illustrate the new methodology with two original data analyses, one regarding microRNA expression in human blastocyst cultures, and another regarding alien plant species invasion in the island of Socotra (Yemen).

We propose a multi-state quantile regression model that admits a cure-fraction for each possible transition, so that individuals may not experience that event. A discrete latent variable allows us to take into account unobserved heterogeneity. The model is estimated in a Bayesian framework, without specification of the number of latent classes. We are motivated by an original application to spells of imprisonment in the USA.

We propose a multi-state quantile regression model that admits a cure-fraction for each possible transition, so that individuals may not experience that event. A discrete latent variable allows us to take into account unobserved heterogeneity. The model is estimated in a Bayesian framework, without specifying the number of latent classes. A simple strategy to scale inference to big data is discussed. We are motivated by an original application to jail recidivism in the U.S. between 2020 and 2023. We find that 20% of the subjects have high cumulative hazard of recidivism; with little association to covariates such as age, gender, crime, and ethnicity. A latent group has been shown to accumulate up to two detentions per year of freedom and represents about 10% of the population.

Background/Objectives: The prognostic value of baseline clinical parameters in predicting the survival prolonging effect of Radium-223-dichloride (223RaCl2) for metastatic castration resistant prostate cancer (mCRPC) patients has been the object of intensive research and remains an open issue. This national multicenter study aimed to corroborate the evidence of ten years of clinical experience with 223RaCl2 by collecting data from eight Italian Nuclear Medicine Units. Methods: Data from 581 consecutive mCRPC patients treated with 223RaCl2 were retrospectively analyzed. Several baseline variables relevant to the overall survival (OS) analysis were considered, including age, previous radical prostatectomy/radiotherapy, number of previous treatment lines, prior chemotherapy, Gleason score, presence of lymphoadenopaties, number of bone metastases, concomitant use of bisphosphonates/Denosumab, Eastern Cooperative Oncology Group Performance Status (ECOG-PS), as well as baseline values of hemoglobin (Hb), platelets, Total Alkaline Phosphatase (tALP), Lactate Dehydrogenase (LDH), and Prostate-Specific Antigen (PSA). Data were summarized using descriptive statistics, univariate analysis and multivariate analysis with the Cox model. Results: The median OS time was 14 months (95%CI 12-17 months). At univariate analysis age, the number of previous treatment lines, number of bone metastases, ECOG-PS, presence of lymphadenopathies at the time of enrollment, as well as baseline tALP, PSA, and Hb, were independently associated with OS. After multivariate analysis, the number of previous treatment lines (HR = 1.1670, CI = 1.0095-1.3491, p = 0.0368), the prior chemotherapy (HR = 0.6461, CI = 0.4372-0.9549, p = 0.0284), the presence of lymphadenopathies (HR = 1.5083, CI = 1.1210-2.0296, p = 0.0066), the number of bone metastases (HR = 0.6990, CI = 0.5416-0.9020, p = 0.0059), ECOG-PS (HR = 1.3551, CI = 1.1238-1.6339, p = 0.0015), and baseline values of tALP (HR = 1.0008, CI = 1.0003-1.0013, p = 0.0016) and PSA (HR = 1.0004, CI = 1.0002-1.0006, p = 0.0005) remained statistically significant. Conclusions: In the era of precision medicine and in the landscape of novel therapies for mCRPC, the prognostic stratification of patients undergoing 223RaCl2 has a fundamental role for clinical decision-making, ranging from treatment choice to optimal sequencing and potential associations. This large Italian multicenter study corroborated the prognostic value of several variables, emerging from ten years of clinical experience with 223RaCl2.

We introduce a novel framework for multivariate time series that demonstrates the powerful synergy between MIxed-DAta Sampling (MIDAS) and Markov-switching models. Specifically, we derive a general multivariate hidden semi-Markov model with configuration of latent states that is not known in advance. Bayesian inference is performed by means of a Reversible Jump Markov Chain Monte Carlo algorithm. We illustrate with a real data application on the association between energy consumption, production, and prices.

We introduce quantile share ratio regression. Our proposal involves the specification of a generalised linear model for the ratio of tail areas above and below two pre-specified quantiles. The latter is the quantile share ratio, a measure of primary interest in the study of income inequality. Our specification is completely distribution-free. We introduce an efficient two-step approach for parameter estimation that entails estimation of the conditional cumulative distribution function at the first step. A scalable strategy is discussed for large sample sizes. We are motivated by the study of income inequality in the European Union, using data from a sample of about three million households.

In the class of hidden semi-Markov models with non-parametric sojourn-time distribution, we present a framework that penalises the latter with respect to its departure from a parametric base kernel. The penalised approach explicitly bridges parametric and non-parametric assumptions for the sojourn-time distributions, also in terms of the effective number of parameters. Inference is obtained via an expectation–maximisation algorithm. For Ridge-type penalties, we reduce the M step to a univariate optimisation problem, thereby greatly improving the computational burden. The penalty parameter is chosen using a computationally efficient Akaike-type information criterion. We illustrate our method with a simulation study, and a real data application to a large number of time series measuring traffic flow in several spots of five European cities. In the real data example, penalised hidden semi-Markov models are often preferred to hidden Markov models, and to hidden semi-Markov models with parametric and non-parametric sojourn-time specifications.

We consider the problem of estimating the conditional quantiles of an unknown distribution from data gathered on a spatial domain. We propose a spatial quantile regression model with differential regularisation. The penalisation involves a partial differential equation defined over the considered spatial domain, that can display a complex geometry. Such regularisation permits, on one hand, to model complex anisotropy and non-stationarity patterns, possibly on the basis of problem-specific knowledge, and, on the other hand, to comply with the complex conformation of the spatial domain. We define an innovative functional Expectation–Maximisation algorithm, to estimate the unknown quantile surface. We moreover describe a suitable discretisation of the estimation problem, and investigate the theoretical properties of the resulting estimator. The performance of the proposed method is assessed by simulation studies, comparing with state-of-the-art techniques for spatial quantile regression. Finally, the considered model is applied to two real data analyses, the first concerning rainfall measurements in Switzerland and the second concerning sea surface conductivity data in the Gulf of Mexico.

Background: The deep inferior epigastric perforator (DIEP) flap is universally considered the gold standard technique for breast reconstruction (BR), though it cannot always be proposed to patients with insufficient donor-site volume. We explore the efficacy of autologous fat transfer (AFT) of the Holm abdomen zone IV in the retropectoral plane during DIEP flap reconstruction (lipo-DIEP flap), to enhance the volume provided by the abdominal donor site in patients with low body mass index (BMI). Methods: We prospectively enrolled patients with BMI less than 25 kg/m2 and candidates for lipo-DIEP flap BR (group A) comparing them with a control group (group B) undergoing traditional DIEP flap BR with the same characteristics of the first group (BMI < 25 kg/m2). Patients belonging to group A underwent magnetic resonance imaging preoperatively and 6 months after the BR, evaluating the adipose tissue volume retained in the retropectoral space. Results: A total of 40 breasts were included in the study. The 2 groups were homogeneous regarding the collected variables, except for mean delayed AFT sessions (0.25 versus 0.95; P= 0.00094). The average volume of retropectoral AFT was 116.25 mL (SD 31.36). Six months after the procedure, the mean retropectoral fat volume calculated through magnetic resonance imaging was 48.64 mL (SD 14.15), whereas the mean graft integration rate was 45.98% (range, 30.7%–64.2%). Conclusions: The lipo-DIEP flap is a valuable technique for patients with insufficient donor-site volume. Immediate retropectoral fat grafting from the Holm zone IV has proven to be safe in terms of complications, reducing the need for further AFT sessions.

Background: Implant-based breast reconstruction (BR) is to date the most popular reconstructive modality. The aim of the study was to evaluate the immediate hybrid one-stage BR technique, with a dual-plane approach, using prepectoral implants and retro-pectoral autologous fat transfer (AFT). Methods: We prospectively enrolled patients scheduled for immediate BR using a hybrid approach, which included retro-pectoral AFT and prepectoral breast implants (Group-A). This cohort was compared with a retrospective control group of patients who underwent immediate direct-to-implant BR without AFT (Group-B). Complications, hospitalization days, number of AFT procedures for a complete BR, aesthetic outcomes and patient’s satisfaction were analysed. Fat survival rate was assessed through comparison of preoperative and 6-months postoperative MRI. Results: 30 immediate BR were included in each group. The average amount of AFT in the retro-pectoral plane was 106.30 cc (SD 16.54), while the mean breast implant size was statistically higher in group B (p=0.00026). MRI assessment confirmed an average of 47.90 % (SD 0.14) retro-pectoral fat survival at 6 postoperative months. No statistically significant differences in term of complications and hospitalization days were observed (p>0.05), while a significant difference was observed regarding additional AFT sessions (p=0.00062). After a propensity score weighted analysis, surgeons and patients assessment showed a significant higher overall satisfaction in the active group. Conclusions: Immediate hybrid approach for one-stage BR, enabled the use of smaller breast implants, a decrease in AFT procedures with costs reduction, and yielded optimal breast shape and contour, by reducing step-off deformity and rippling alterations, without adding complications. Level of Evidence II