On confidence intervals for semiparametric expectile regression

Abstract

In regression scenarios there is a growing demand for information on the conditional distribution of the response beyond the mean. In this scenario quantile regression is an established method of tail analysis. It is well understood in terms of asymptotic properties and estimation quality. Another way to look at the tail of a distribution is via expectiles. They provide a valuable alternative since they come with a combination of preferable attributes. The easy weighted least squares estimation of expectiles and the quadratic penalties often used in flexible regression models are natural partners. Also, in a similar way as quantiles can be seen as a generalisation of median regression, expectiles offer a generalisation of mean regression. In addition to regression estimates, confidence intervals are essential for interpretational purposes and to assess the variability of the estimate, but there is a lack of knowledge regarding the asymptotic properties of a semiparametric expectile regression estimate. Therefore confidence intervals for expectiles based on an asymptotic normal distribution are introduced. Their properties are investigated by a simulation study and compared to a boostrap-based gold standard method. Finally the introduced confidence intervals help to evaluate a geoadditive expectile regression model on childhood malnutrition data from India.

An SVM-like approach for expectile regression

“…The clipped slack variables used in the stopping criteria (27) may provide a substantial decrease in duality gap in each iteration of learning algorithm compared to the unclipped slack variables used in (21), and hence the learning algorithm may require less number of iterations. [34] showed that the right hand side of the stopping criteria given in (21) should be replaced by 2λ as in (27), where has the same value for both.…”

A location‐scale model for non‐crossing expectile curves

“…A package for R, expectreg, is available in Sobotka et al (2013). It can fit the simple LAWS model (Schnabel & Eilers, 2009b) well as the expectile bundle described in this manuscript.…”

Change point and trend analyses of annual expectile curves of tropical storms

“…It combined the LAWS (least average weighted squares) algorithm with P-splines in order to estimate expectile curves. Recent applications include Guo and Härdle (2012), Sobotka et al (2013) and Guo et al (2015) or more applicable one in finance by Taylor (2008), where Value at risk (VaR) and Expected shortfall (ES) were estimated using expectiles.…”