Very Preterm and Full‐Term Infants’ Response to Socio‐Emotional Stress: The Role of Postnatal Maternal Bonding

Very preterm (VPT) infants are at risk for altered socio‐emotional stress regulation, even in the absence of perinatal complications. Moreover, becoming a mother of a VPT infant is a traumatic event, which might affect the establishment of a bonding relationship toward the infant during the postnatal period. The main aim of the present study was to assess the association between maternal postnatal bonding and socio‐emotional stress response in 3‐month‐old VPT infants, compared to full‐term (FT) controls. Thirty‐three VPT and 28 FT infant‐mother dyads underwent the maternal Face‐to‐Face Still‐Face (FFSF) procedure to assess socio‐emotional stress regulation. Maternal postnatal bonding was measured through the Maternal Post‐Natal Attachment Scale (MPAS), which includes three scales: Quality of attachment, Absence of hostility, and Pleasure in interaction. The VPT infants exhibited less self‐directed and other‐directed regulatory behaviors during the Still‐Face episode. Mothers of VPT infants reported lower Quality of attachment compared to FT counterparts. Importantly, higher levels of Quality of attachment were associated with more other‐directed behaviors during the Still‐Face episode in FT dyads, whereas this was not observed in VPT dyads. In conclusion, VPT infants might be exposed to a double‐risk condition for socio‐emotional development, encompassing both difficulties in stress response and the exposure to a less‐than‐optimal maternal bonding.     

Adjusting Stepwise p-Values in Generalized Linear Models

Stepwise methods for variable selection are frequently used to determine the predictors of an outcome in generalized linear models. Although it is widely used within the scientific community, it is well known that the tests on the explained deviance of the selected model are biased. This arises from the fact that the traditional test statistics upon which these methods are based were intended for testing pre-specified hypotheses; instead, the tested model is selected through a data-driven procedure. A multiplicity problem therefore arises. In this work, we define and discuss a nonparametric procedure to adjust the p-value of the selected model of any stepwise selection method. The unbiasedness and consistency of the method is also proved. A simulation study shows the validity of this procedure. Theoretical differences with previous works in the same field are also discussed.     

FDR- and FWE-controlling methods using data-driven weights

Weighted methods are an important feature of multiplicity control methods. The weights must usually be chosen a priori, on the basis of experimental hypotheses. Under some conditions, however, they can be chosen making use of information from the data (therefore a posteriori) while maintaining multiplicity control. In this paper we provide: (1) a review of weighted methods for familywise type I error rate (FWE) (both parametric and nonparametric) and false discovery rate (FDR) control; (2) a review of data-driven weighted methods for FWE control; (3) a new proposal for weighted FDR control (data-driven weights) under independence among variables; (4) under any type of dependence; (5) a simulation study that assesses the performance of procedure of point 4 under various conditions.     

Neurocognitive Effects in Welders Exposed to Aluminium: An Application of the NPC Test and NPC Ranking Methods

The aim of this study is to investigate the effects of exposure to aluminium on the cognitive sphere in a group of welders. Given the intrinsic complexity of the neurocognitive sphere, three different psychomotor variables were used for its investigation allowing the resulting problem to be naturally modelized into a multivariate framework and solved by a nonparametric combination of permutation tests (NPC test). In order to make the treated and control groups comparable, we also stratified the samples on the basis of a risk index that estimates the combined action of biological age and exposure time. Despite the fact that studies in the literature have reached conflicting results, our study highlights a significant drop in attention and memory performances in individuals exposed to aluminium. Finally, we identify a global criterion that summarises the information on the neurocognitive state by applying the nonparametric combination of dependent rankings method (NPC ranking).     

A Permutation Approach to Split-Plot Experiments

Split-plot design may be refer to a common experimental setting where a particular type of restricted randomization has occurred during a planned experiment. The aim of this article is to suggest a new method to perform inference on split-plot experiments by combination-based permutation tests. This novel nonparametric approach has been studied and validated using a Monte Carlo simulation study where we compared it with the parametric and nonparametric procedures proposed in the literature. Results suggest that in each experimental situation where normality is hard to justify and especially when errors have heavy-tailed distribution, the proposed nonparametric procedure can be considered as a valid solution.     

Forecasting composite indicators with anticipated information: an application to the industrial production index

Summary. Many economic and social phenomena are measured by composite indicators computed as weighted averages of a set of elementary time series. Often data are collected by means of large sample surveys, and processing takes a long time, whereas the values of some elementary component series may be available a considerable time before the others and may be used for forecasting the composite index. This problem is addressed within the framework of prediction theory for stochastic processes. A method is proposed for exploiting anticipated information to minimize the mean-square forecast error, and for selecting the most useful elementary series. An application to the Italian general industrial production index is illustrated, which demonstrates that knowledge of anticipated values of some, or even just one, component series may reduce the forecast error considerably.     

A Critical Review and a Comparative Study on Conditional Permutation Tests for Two-Way ANOVA

Two-way ANOVA methodology is surely one of the most important models in the framework of the experimental design theory, as suggested by the great number of proposed solutions given in literature. Among these, some solutions are nonparametric and particularly, thanks to the availability of modern powerful computing equipments, those based on conditional on observations permutation test have gained great interest. The aim of this work is to present and compare such proposals and to illustrate their possible advantages and disadvantages when applied to some real data-sets.     

Permutation tests for between-unit fixed effects in multivariate generalized linear mixed models

A permutation testing approach in multivariate mixed models is presented. The solutions proposed allow for testing between-unit effect; they are exact under some assumptions, while approximated in the more general case. The classes of models comprised by this approach include generalized linear models, vector generalized additive models and other nonparametric models based on smoothing. Moreover it does not assume observations of different units to have the same distribution. The extensions to a multivariate framework are presented and discussed. The proposed multivariate tests exploit the dependence among variables, hence increasing the power with respect to other standard solutions (e.g. Bonferroni correction) which combine many univariate tests in an overall one. Examples are given of two applications to real data from psychological and ecological studies; a simulation study provides some insight into the unbiasedness of the tests and their power. The methods were implemented in the R package flip, freely available on CRAN.     

A Discussion on Nonlinear Models for Price Decisions in Rating-Based Product Preference Models

The purpose of this article is to discuss the application of nonlinear models to price decisions in the framework of rating-based product preference models. As revealed by a comparative simulation study, when a nonlinear model is the true model, the traditional linear model fails to properly describe the true pattern. It appears to be unsatisfactory in comparison with nonlinear models, such as logistic and natural spline, which offer some advantages, the most important being the ability to take into account more than just linear and/or monotonic effects. Consequently, when we model the product preference with a nonlinear model, we are potentially able to detect its ‘best’ price level, i.e., the price at which consumer preference towards a given attribute is at its maximum. From an application point of view, this approach is very flexible in price decisions and may produce original managerial suggestions which might not be revealed by traditional methods.     

LASSO order selection for sparse autoregression: a bootstrap approach

Autoregressive models are widely employed for predictions and other inferences in many scientific fields. While the determination of their order is in general a difficult and critical step, this task becomes more complicated and crucial when the time series under investigation is realization of a stochastic process characterized by sparsity. In this paper we present a method for order determination of a stationary AR model with a sparse structure, given a set of observations, based upon a bootstrapped version of MAICE procedure [Akaike H. Prediction and entropy. Springer; 1998], in conjunction with a LASSO-type constraining procedure for lag suppression of insignificant lags. Empirical results will be obtained via Monte Carlo simulations. The quality of our method is assessed by comparison with the commonly adopted cross-validation approach and the non bootstrap counterpart of the presented procedure.