1: \begin{abstract}

2: \noindent We consider method-of-quantiles estimators of unknown

3: parameters, namely the analogue of method-of-moments estimators

4: obtained by matching empirical and theoretical quantiles at some

5: probability level $\lambda\in(0,1)$. The aim is to present large

6: deviation results for these estimators as the sample size tends to

7: infinity. We study in detail several examples; for specific models

8: we discuss the choice of the optimal value of $\lambda$ and we

9: compare the convergence of the method-of-quantiles and

10: method-of-moments estimators.\\

11: \ \\

12: \emph{AMS Subject Classification:} 60F10; 62F10; 62F12.\\

13: \emph{Keywords:} location parameter; methods of moments; order

14: statistics; scale parameter; skewness parameter.

15: \end{abstract}

16: