---

Titolo: Parametric approaches to quantile regression

Abstract: I present a method for simultaneous quantile regression, in which
coefficients are modelled by parametric functions. Compared to
ordinary quantile regression, in which different quantiles are
estimated one at a time, the proposed approach presents various
advantages in terms of computation, efficiency, and interpretation of
the results, and allows for particularly convenient extensions to
survival data (e.g., censored and truncated data), longitudinal data,
and count data. Moreover, it makes it simple to handle quantile
crossing, perform covariates selection, and estimate extremes. The
estimator is implemented in the R packages qrcm, qrcmPen, and Qest.