Quantile regression can be used to obtain a non-parametric estimate of aconditional quantile function. The presence of quantile crossing, however, leads toan invalid distribution of the response and makes it difficult to use the fitted modelfor prediction. In this work, we show that crossing can be alleviated by modellingthe quantile function parametrically. We then describe an algorithm for constrainedoptimisation that can be used to estimate parametric quantile functions with the noncrossingproperty. We investigate climate change by modelling the long-term trendsof extreme temperatures in the Arctic Circle.
|Title of host publication||Smart Statistics for Smart Applications - Book of Short Papers SIS2019|
|Number of pages||8|
|Publication status||Published - 2019|