In questo contributo vogliamo discutere di un modello di regressione segmented per valutare e qua... more In questo contributo vogliamo discutere di un modello di regressione segmented per valutare e quantificare l\u2019effetto delle misure di contenimento individuando gli istanti temporali in cui la diffusione della malattia rallenta
We propose a new lasso-type estimator of regression coefficients for regression models. Our propo... more 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
This paper focuses on interval estimation in logistic regression models fitted through the Firth ... more 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 penalized logistic regression
We propose an iterative algorithm to select the smoothing parameters in additive quantile regress... more We propose an iterative algorithm to select the smoothing parameters in additive quantile regression, wherein the functional forms of the covariate effects are unspecified and expressed via B-spline bases with difference penalties on the spline coefficients. The proposed algorithm relies on viewing the penalized coefficients as random effects from the symmetric Laplace distribution, and it turns out to be very efficient and particularly attractive with multiple smooth terms. Through simulations we compare our proposal with some alternative approaches, including the traditional ones based on minimization of the Schwarz Information Criterion. A real-data analysis is presented to illustrate the method in practice.
Segmented mixed models with random changepoints: a maximum likelihood approach with application t... more Segmented mixed models with random changepoints: a maximum likelihood approach with application to treatment for depression study
A core task in analyzing randomized clinical trials based on lon- gitudinal data is to nd the bes... more A core task in analyzing randomized clinical trials based on lon- gitudinal data is to nd the best way to describe the change over time for each treatment arm. We review the implementation and estimation of a exible piecewise Hierarchical Linear Model (HLM) to model change over time. The
Background: Asthma patterns are not well established in epidemiological studies. Aim: To assess a... more Background: Asthma patterns are not well established in epidemiological studies. Aim: To assess asthma patterns and risk factors in an adult general population sample. Methods: In total, 452 individuals reporting asthma symptoms/diagnosis in previous surveys participated in the AGAVE survey (2011–2014). Latent transition analysis (LTA) was performed to detect baseline and 12-month follow-up asthma phenotypes and longitudinal patterns. Risk factors associated with longitudinal patterns were assessed through multinomial logistic regression. Results: LTA detected four longitudinal patterns: persistent asthma diagnosis with symptoms, 27.2%; persistent asthma diagnosis without symptoms, 4.6%; persistent asthma symptoms without diagnosis, 44.0%; and ex -asthma, 24.1%. The longitudinal patterns were differently associated with asthma comorbidities. Persistent asthma diagnosis with symptoms showed associations with passive smoke (OR 2.64, 95% CI 1.10–6.33) and traffic exposure (OR 1.86, 95%...
In questo contributo vogliamo discutere di un modello di regressione segmented per valutare e qua... more In questo contributo vogliamo discutere di un modello di regressione segmented per valutare e quantificare l\u2019effetto delle misure di contenimento individuando gli istanti temporali in cui la diffusione della malattia rallenta
We propose a new lasso-type estimator of regression coefficients for regression models. Our propo... more 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
This paper focuses on interval estimation in logistic regression models fitted through the Firth ... more 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 penalized logistic regression
We propose an iterative algorithm to select the smoothing parameters in additive quantile regress... more We propose an iterative algorithm to select the smoothing parameters in additive quantile regression, wherein the functional forms of the covariate effects are unspecified and expressed via B-spline bases with difference penalties on the spline coefficients. The proposed algorithm relies on viewing the penalized coefficients as random effects from the symmetric Laplace distribution, and it turns out to be very efficient and particularly attractive with multiple smooth terms. Through simulations we compare our proposal with some alternative approaches, including the traditional ones based on minimization of the Schwarz Information Criterion. A real-data analysis is presented to illustrate the method in practice.
Segmented mixed models with random changepoints: a maximum likelihood approach with application t... more Segmented mixed models with random changepoints: a maximum likelihood approach with application to treatment for depression study
A core task in analyzing randomized clinical trials based on lon- gitudinal data is to nd the bes... more A core task in analyzing randomized clinical trials based on lon- gitudinal data is to nd the best way to describe the change over time for each treatment arm. We review the implementation and estimation of a exible piecewise Hierarchical Linear Model (HLM) to model change over time. The
Background: Asthma patterns are not well established in epidemiological studies. Aim: To assess a... more Background: Asthma patterns are not well established in epidemiological studies. Aim: To assess asthma patterns and risk factors in an adult general population sample. Methods: In total, 452 individuals reporting asthma symptoms/diagnosis in previous surveys participated in the AGAVE survey (2011–2014). Latent transition analysis (LTA) was performed to detect baseline and 12-month follow-up asthma phenotypes and longitudinal patterns. Risk factors associated with longitudinal patterns were assessed through multinomial logistic regression. Results: LTA detected four longitudinal patterns: persistent asthma diagnosis with symptoms, 27.2%; persistent asthma diagnosis without symptoms, 4.6%; persistent asthma symptoms without diagnosis, 44.0%; and ex -asthma, 24.1%. The longitudinal patterns were differently associated with asthma comorbidities. Persistent asthma diagnosis with symptoms showed associations with passive smoke (OR 2.64, 95% CI 1.10–6.33) and traffic exposure (OR 1.86, 95%...
Uploads
Papers by Vito Muggeo