Improving the calibration of the best member method using quantile regression to forecast extreme temperatures

Abstract

Temperature influences both the demand and supply of electricity and is therefore a potential cause of blackouts. Like any electricity provider, Electricité de France (EDF) has strong incentives to model the uncertainty in future temperatures using ensemble prediction systems (EPSs). However, the probabilistic representations of the future temperatures provided by EPSs are not reliable enough for electricity generation management. This lack of reliability becomes crucial for extreme temperatures, as these extreme temperatures can result in blackouts. A proven method to solve this problem is the best member method (BMM). This method improves the representation as a whole, but there is still room for improvement in the tails of the distribution. The idea of the BMM is to model the probability distribution of the difference between the forecast and realization. We improve the error modeling in BMM using quantile regression, which is more efficient than the usual two-stage ordinary least squares (OLS) regression. To achieve further improvement, the probability that a given forecast is the best one can be modeled using exogenous variables

Effective analysis of job satisfaction among medical staff in Chinese public hospitals: a random forest model

“…Multiple logistic regression models are widely used in related research due to their relatively simple theoretical assumptions ( 26 ). Regression models have greater advantages over OLS models in probability prediction ( 27 ). Li et al analyzed the influencing factors of patient satisfaction through a multiple logistic regression model ( 28 ).…”

A Generic Method for Density Forecasts Recalibration

“…Nevertheless, one can state that there is no formal test of calibration in these works. In [3], the only measure used is the CRPS, and [8] addresses exclusively the calibration issue.…”