Search results for "Quantile"
showing 10 items of 107 documents
Non-crossing parametric quantile functions: an application to extreme temperatures
2019
Quantile regression can be used to obtain a non-parametric estimate of a conditional quantile function. The presence of quantile crossing, however, leads to an invalid distribution of the response and makes it difficult to use the fitted model for prediction. In this work, we show that crossing can be alleviated by modelling the quantile function parametrically. We then describe an algorithm for constrained optimisation that can be used to estimate parametric quantile functions with the noncrossing property. We investigate climate change by modelling the long-term trends of extreme temperatures in the Arctic Circle.
Poster Presentations
2014
Administration of recombinant human erythropoietin (rHuEpo) improves performance and hence is subject to abuse by athletes. The detection of rHuEpo doping remains a major challenge at present. The aim of the current study was to investigate whether circulating microRNA (miRNA) can be used for detecting r-HuEpo doping. Twenty trained males received rHuEpo injections of 50 IU∙kg -1 body mass every two days for 4 weeks. Blood was obtained 2 weeks before, during and 4 weeks after administration. For this pilot analysis, plasma miRNA expression was assessed at selected time points using the Affymetrix GeneChip 3.0 and the miScript 384 HC PCR Array (Qiagen). For the Affymetrix microarray data, GC…
A new indicator for higher education student performance
2014
The debate on academic achievement is a heated issue that involves all the higher education contexts. This paper attempts to provide an indicator that can make the measurement of university student performance easier and that can be easily applied to different systems, making comparisons more fair. The Italian University System is used as a starting point to make several considerations on the current measures and to build up a new performance indicator. Then, a generalization for other marking systems is shown and finally a quantile regression is performed to investigate some determinants of the new performance indicator, also with respect to the current one.
The Somatic Symptom Disorder - B Criteria Scale (SSD-12): Factorial structure, validity and population-based norms.
2017
The Somatic Symptom Disorder - B Criteria Scale (SSD-12) assesses the psychological features of DSM-5 Somatic Symptom Disorder (SSD). The present study investigates the dimensionality and psychometric properties in a general population sample and provides norm values.Test dimensionality was evaluated via confirmatory factor analysis and nonparametric item response theory. Correlational analyses and logistic regression models based on related measures (SSS 8, PHQ-2, GAD-2, Health Care Utilization) were used to derive predictive validity. Age and gender specific norms were derived via quantile regression.The SSD-12 has good item characteristics and excellent reliability (Cronbach's α=0.95). C…
Individuazione dei trend di pioggia per la Sicilia utilizzando un approccio multiscala basato sulla regressione a quantili
2021
Uno degli argomenti più attuali e controversi è se, negli ultimi decenni, le piogge intense siano diventate più frequenti come possibile effetto dei cambiamenti climatici. La centralità del tema è giustificata dalle gravi conseguenze che l’intensificarsi di questi fenomeni, in frequenza e magnitudo, potrebbe indurre a livello sociale ed economico. Basti pensare alle “flash flood”, spesso causate da eventi di precipitazione con una forte componente convettiva, che in ambito urbano mettono in crisi i sistemi di drenaggio arrecando rilevanti danni e, purtroppo, anche perdite di vite umane. La Sicilia è stata spesso monitorata con l’obiettivo di individuare delle mutazioni significative del cic…
On using novel “Anti-Bayesian” techniques for the classification of dynamical data streams
2017
The classification of dynamical data streams is among the most complex problems encountered in classification. This is, firstly, because the distribution of the data streams is non-stationary, and it changes without any prior “warning”. Secondly, the manner in which it changes is also unknown. Thirdly, and more interestingly, the model operates with the assumption that the correct classes of previously-classified patterns become available at a juncture after their appearance. This paper pioneers the use of unreported novel schemes that can classify such dynamical data streams by invoking the recently-introduced “Anti-Bayesian” (AB) techniques. Contrary to the Bayesian paradigm, that compare…
The student talent in a random effects Quantile Regression Model for university performance
2013
This paper is a development of a previous work on performance of Italian university students. A Quantile Regression was carried out on a proposed performance indicator, based on a transformation of the median of the marks weighted by credits. Results suggested to investigate the role of students peculiar features on their performance, measured by the transformation of the weighted marks. Therefore, a random intercept Quantile Regression model is fitted on real data concerning graduates over the legal university duration set in Italy, enrolled in 2002 in two Degree Courses of the University of Palermo. Results show that the student performance seems to be influenced by the student's aptitude…
Do personal characteristics affect the Rasch measures of perceived physical risk? A quantile regression approach
2012
This paper focuses on the measurement of the perception of healthy risk according to personal characteristics. The Physical Risk Assessment Inventory was adopted as measurement tool and it was administered to 551 students enrolled in the first and in the fifth classes of some high schools of Palermo. The analysis of the determinants of the perceived risk is based on its quantitative measures. Therefore the analysis has been developed into two tracks. First track is devoted to obtain a quantitative measure of the perceived risk: an Extended Logistic Rasch Model was used considering separately males and females. Results highlight the different perception of risk between males and females, alt…
“Anti-Bayesian” flat and hierarchical clustering using symmetric quantiloids
2017
A myriad of works has been published for achieving data clustering based on the Bayesian paradigm, where the clustering sometimes resorts to Naive-Bayes decisions. Within the domain of clustering, the Bayesian principle corresponds to assigning the unlabelled samples to the cluster whose mean (or centroid) is the closest. Recently, Oommen and his co-authors have proposed a novel, counter-intuitive and pioneering PR scheme that is radically opposed to the Bayesian principle. The rational for this paradigm, referred to as the “Anti-Bayesian” (AB) paradigm, involves classification based on the non-central quantiles of the distributions. The first-reported work to achieve clustering using the A…
Assessing changes in rainfall erosivity in Sicily during the twentieth century
2007
Changes in rainfall erosivity are an expected consequence of climate change. Long-term series of the single storm erosion index, EI, may be analysed to detect trends in rainfall erosivity. An indirect approach has to be applied for estimating EI, given that long series of rainfall intensities are seldom available. In this paper, a method for estimating EI from the corresponding rainfall amount, he, was developed for Sicily. This method was then applied at 17 Sicilian locations, representative of different climatic zones of the region, to generate a long series (i.e. from 1916 to 1999 in most cases) of EI values. Linear and step (step located at 1970) trends in annual and seasonal erosivity …