Growth curves of sorghum roots via quantile regression with P-splines

Plant roots are a major pool of total carbon in the planet and their dynamics are directly relevant to greenhouse gas balance. Composted wastes are increasingly used in agriculture for environmental and economic reasons and their role as a substitute for traditional fertilizers needs to be tested on all plant components. Here we propose a regression quantile approach based on P-splines to assess, quantify and compare the root growth patterns in two treatment groups respectively undergoing compost and traditional fertilization.

LASSO regression via smooth L1-norm approximation

This paper discusses estimation of regression model with LASSO penalty when the L1-norm is replaced with its parametric smooth approximation. The resulting parameter estimators are more manageable than those from standard LASSO, standard errors are easy computed via a sandwich formula, and the model degrees of freedom may be computed straightforwardly. Moreover the resulting objective function may be minimized using usual optimization algorithms for regular models, for instance Newton-Raphson or iterative least squares.

Using Zero-Inflated Models to analyze environmental data sets with many zeroes

Altered Yin Yang 1/RAF-1 kinase inhibitory protein ratio as a possible molecular marker and therapeutic target in hepatocellular carcinoma

Approximate piecewise linear mixed modelling with random changepoints for longitudinal data analysis

La stima del crossing-point nel modello di Cox per l'analisi di sopravvivenza: una proposta

La variazione dell'effetto di un fattore di rischio in un modello di sopravvivenza a rischi proporzionali di Cox è di particolare interesse in molte applicazioni, in quanto l'allontanamento dall'ipotesi di proporzionalità dei rischi rende il modello inadeguato. Di particolare interesse è il caso in cui le funzioni di rischio di due gruppi si intersecano, ovvero, più in generale, il coefficiente di regressione cambia segno dopo un istante temporale detto crossing-point. In questo lavoro si propone una procedura per la stima del crossing-point applicabile anche al caso di variabili quantitative, e si discutono tre diversi metodi per la corrispondente stima intervallare. L'approccio è illustra…

Penalized logistic regression for small or sparse data: interval estimators revisited

This paper focuses on interval estimation in logistic regression models fitted through the Firth penalized log-likelihood. In this context, many authors have claimed superiority of the Likelihood ratio statistic with respect to the (wrong) Wald statistic via simulation evidence. We re-assess such findings by detailing the inferential tools also including in the comparisons the (right) Wald statistic and other statistics neglected in previous literature. In particular, we assess performances of the CIs estimators by simulation and compare them in a real data set. Differently from previous findings, the Likelihood ratio statistic does not appear to be the best inferential device in Firth pena…

Assessing and summarizing temperature effects in native and migrant residents in Torino, 1972-2002

Comment on "Estimating average annual per cent change in trend analysis"

We discuss some issues relevant to paper of Clegg and co-authors published in Statistics in Medicine; 28, 3670-3682. Emphasis is on computation of the variance of the sum of products of two estimates, slopes and breakpoints.

A note on temperature effect estimate in mortality time series analysis

Assessment of Cardiovascular Function in Childhood Leukemia Survivors: The Role of the Right Heart

Childhood acute lymphoblastic leukemia (ALL) survivors who underwent chemotherapy with anthracyclines have an increased cardiovascular risk. The aim of the study was to evaluate left and right cardiac chamber performances and vascular endothelial function in childhood ALL survivors. Fifty-four ALL survivors and 37 healthy controls were enrolled. All patients underwent auxological evaluation, blood pressure measurements, biochemical parameters of endothelial dysfunction, flow-mediated dilatation (FMD) of the brachial artery, mean common carotid intima-media thickness (c-IMT), antero-posterior diameter of the infra-renal abdominal aorta (APAO), and echocardiographic assessment. The ALL subjec…

Testing for a breakpoint in segmented regression: a pseudo score approach

To overcome the well known oddities in testing for the existence of a breakpoint in segmented regression models, we discuss a novel approach based on the Pearson X2 statistic which can be understood as an approximation of the Score statistic. We describe the method and present results from some simulations.

segmented: An R package to Fit Regression Models with Broken-Line Relationships

Segmented or broken-line models are regression models where the relationships between the response and one or more explanatory variables are piecewise linear, namely represented by two or more straight lines connected at unknown values: these values are usually referred as breakpoints, changepoints or even joinpoints.

DOES AIR POLLUTION MODIFY THE HEAT TOLERANCE? ESTIMATING THE THRESHOLD-LINE VIA SEGMENTED REGRESSION

Influenza del substrato su crescita dei rizomi e biometria fogliare di Posidonia oceanica

Effect of temperature on mortality in Torino: the role of birthplace - preliminary analyses 1980-89

Using ZIP models to analyse environmental time series with many zeroes

The induced smoothed LASSO

We propose a new lasso-type estimator of regression coefficients for regression models. Our proposal relies on the recent idea of induced smoothing and leads to estimators with sampling distribution somewhat close to the Normal one, regardless of their true value, along with the corresponding reliable covariance matrix. As a consequence inference (e.g. p-values) may be carried out relatively easily. We present results from some simulation experiments.

Estimation of linear errors-in-variable models with error free covariates: a backfitting approach

We present a backfitting algorithm to estimate linear regression mod- els having both error-prone and error-free covariates as predictors. The algorithm assumes that the variance-ratios are known, and it is particulary efficient when several explanatory variables are included. The resulting estimators are shown to be unbiased and to perform well as compared to method-of-moments estimators which are usually employed when the variance ratio is known.

Change-point estimation in piecewise constant regression models with random effects

We propose an iterative algorithm to estimate change-points in general regression models. The algorithm avoids grid search to obtain maximum likelihood estimates, and thus it guarantees moderate computational time regardless of the sample size and the number of change-points to be estimated. Furthermore, it allows estimation in random effects models, where grid search is unfeasible. We present the proposed approach in practice by analyzing variations of lung functionality on a sample of transplant recipients.

Estimation of GLMs with unspecified link function: a P-spline approach

Modelling the timing of divorce in Italy: a survival analysis on regression quantiles

The analysis of marital dissolution in Italy represents a quite interesting and challenging topic from a substantive standpoint; in fact, despite of the decreasing number of marriages and the increasing number of divorces, the traditional family based on the marriage of heterosexual partners is still considered as a fundamental institution of the society. Here we present a censored quantile regression model with additive terms to investigate the determinants of the timing of marital dissolution on a large and substantial sample from a survey carried on in Italy.

Curve di crescita di riferimento nel monitoraggio di Posidonia oceanica: alcune stime preliminari

A model for the analysis of the temperature effects on mortality

Modelling threshold parameter as a function of covariates in segmented regression

Effetto a breve termine dell'inquinamento sulla salute: Palermo 1997-2002

Factors that cause University students to drop out. An alternative modelling of interaction terms in logistic regression models

Smoothed score confidence interval for the breakpoint in segmented regression

For the breakpoint parameter in segmented regression we consider confidence intervals based on the score statistic. Due to unsmoothness of the score, we propose to build the confidence intervals using its smoothed version under proper shape restrictions. Some simulations are presented to assess the finite sample performance of the proposed approach.

Fitting mean-shift models with covariates