ABSTRACT The tail of a probability distribution captures important information about the magnitude and frequency extreme events. Thus, understanding behavior is crucial in decision making risk assessment. Statistical analysis predictions events often require estimation value index, key parameter that describes heaviness. We present new class estimators positive based on weighted moment method, flexible framework for inference. asymptotic normality introduced established, comparison with Hill...

In Statistics of Extremes, the estimation parameters extreme or even rare events is usually done under a semi-parametric framework. The estimators are based on largest k-ordered statistics in sample excesses over high level u. Although showing good asymptotic properties, most those present strong dependence k u with bias when increases decreases. use resampling methodologies has revealed to be promising reduction and choice Different approaches for need considered depending whether we an...

Views Icon Article contents Figures & tables Video Audio Supplementary Data Peer Review Share Twitter Facebook Reddit LinkedIn Tools Reprints and Permissions Cite Search Site Citation Frederico Caeiro, Dora Prata Gomes; Adaptive estimation of a tail shape second order parameter: A computational comparative study. AIP Conf. Proc. 31 December 2015; 1702 (1): 030005. https://doi.org/10.1063/1.4938771 Download citation file: Ris (Zotero) Reference Manager EasyBib Bookends Mendeley Papers EndNote...

In extreme value theory the extremal index is a key parameter that enables straightforward extension of classic results for independent case to stationary processes, measuring degree local dependence in largest observations. Its estimation important not only by itself but also because its effect on other parameters events. The estimators considered literature, despite having good asymptotic properties, show strong high level un, presenting variance levels and bias when decreases. It has been...

Chromosomal fragile sites (CFSs) are loci or regions susceptible to spontaneous induced occurrence of breaks and rearrangements. They classified in two main categories, rare common, depending on their frequency the population. In order identify which CFSs influential significant deletions duplications (chromosomal constitutional imbalances), we propose a logistic regression analysis for CFS data set, since underlying response variable is categorical, specifically binary (deletion...

Natural hazards such as high rainfall and windstorms arise due to physical processes are usually spatial in its nature. Classical geostatistics, mostly based on multivariate normal distributions, is inappropriate for modeling tail behavior. Several methods have been proposed the of extremes, among which max-stable perhaps most well known. They form a natural class extending extreme value theory when sample maxima observed at each site process. Jointly with theoretical framework...

Floods are recurrent events that can have a catastrophic impact. In this work we interested in the analysis of data set gauged daily flows from Whiteadder Water river, Scotland. Using statistic techniques based on extreme value theory, estimate several parameters, including quantiles and return periods high levels.

Human chromosomal fragile sites (CFSs) are heritable loci or regions of the human chromosomes prone to exhibit gaps, breaks and rearrangements. Determining frequency deletions duplications in CFSs may contribute explain occurrence disease due those In this study we analyzed each CFS. Statistical methods, namely data display, descriptive statistics linear regression analysis were applied analyze dataset. We found that FRA15C, FRA16A FRAXB most frequently involved occurring genome.

When modeling extreme events there are a few primordial parameters, among which we refer the value index and extremal index. The measures right tail-weight of underlying distribution characterizes degree local dependence in extremes stationary sequence. Most semi-parametric estimators these parameters show same type behaviour: nice asymptotic properties, but high variance for small values k, number upper order statistics to be used estimation, bias large k. This shows real need choice...

Resampling methodologies have revealed recently as important tools in semi-parametric estimation of parameters the field extremes. Among a few interest, we are here interested extremal index, measure degree local dependence extremes stationary sequence. Most estimators this parameter show same type behavior: nice asymptotic properties but high variance for small values k, number upper order statistics used and bias large k. Two index considered: classical one reduced-bias generalized...

In this paper we consider the semi-parametric estimation of extreme quantiles a right heavy-tail model. We propose new Log Probability Weighted Moment estimator for quantiles, which is obtained from estimators shape and scale parameters tail. Under second-order regular variation condition on tail, underlying distribution function, deduce non degenerate asymptotic behaviour under study present an comparison at their optimal levels. addition, performance illustrated through application to real data.

In Extreme Value Analysis there are a few parameters of particular interest among which we refer to the extremal index, measure extreme events clustering. It is great for initial dependent samples, common situation in many practical situations. Most semi-parametric estimators this parameter show same behavior: nice asymptotic properties but high variance small values k, number upper order statistics used estimation and bias large k. The Mean Square Error, that encompasses variance, usually...