Loss-based approach to two-piece location-scaledistributions with applications to dependent data

Two-piece location-scale models are used for modeling data presenting departuresfrom symmetry. In this paper, we propose an objective Bayesian methodology forthe tail parameter of two particular distributions of the above family: the skewedexponential power distribution and the skewed generalised logistic distribution. Weapply the proposed objective approach to time series models and linear regressionmodels where the error terms follow the distributions object of study. The performanceof the proposed approach is illustrated through simulation experiments and real dataanalysis. The methodology yields improvements in density forecasts, as shown by theanalysis we carry out on the electricity prices in Nordpool markets.