2016•arXiv (Cornell University)Open access

Large deviations for method-of-quantiles estimators of one-dimensional\n parameters

Valeria Bignozzi, Claudio Macci, Lea Petrella

Abstract

We consider method-of-quantiles estimators of unknown parameters, namely the\nanalogue of method-of-moments estimators obtained by matching empirical and\ntheoretical quantiles at some probability level lambda in (0,1). The aim is to\npresent large deviation results for these estimators as the sample size tends\nto infinity. We study in detail several examples; for specific models we\ndiscuss the choice of the optimal value of lambda and we compare the\nconvergence of the method-of-quantiles and method-of-moments estimators.\n

Open-access reader

About this research paper

What this paper is about

Why it matters

A significance statement is not available in the OpenAlex record.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

Key concepts: Quantile, Estimator, Mathematics, Applied mathematics, Convergence (economics), Extremum estimator, Matching (statistics), Statistics