Variable selectionThis paper is concerned with identifying important features in high dimensional data analysis, especially when there are complex relationships among predictors. Without any specication of an actual model, we rst introduce a multiple testing procedure based on the quantile correlation to ...
The number of iterations of the Sequential MCMC sampler at each time \(m_{t}\) is determined based on the cross-chain correlation (see Guhaniyogi et al., 2018). Specifically, we set the number of iterations at time t, denoted as \(m_{t}\), to be the smallest integer s such that...
On the other hand, as a direct application, based on the new communication-efficient distributed CQR and the smooth-threshold estimating equations (Ueki 2009), we further propose a variable selection procedure for the massive data, which can also be realized on the first machine. Theoretically, ...
We attribute the excellent modelling capabilities, especially the very high coefficients of determination and cross- validation correlations to the models' ability to avoid the curse of dimensionality while retaining great flexibility in the regression function9. In addition,12 concurred w ith...
Algorithm 1 and we introduce the extreme conditional quantile estimator. A general issue with gradient boosting is selection of hyperparameter such as the number of trees and the parameters governing the tree structure. We propose adaptive hyperparameters selection with deviance-based cross-validation. ...
However, the strong positive correlation between the two measures in the sample suggests that such a strategy of increasing the Out-degree and decreasing the In-degree may be hard to implement in practice, and that economic dimensions other than EP may impact on the trading intensity. ...
It produces a graphical map based on the forgoing statistical metrics that present the distance between the forecasting models and the observed river flow data set. From the Taylor diagrams, all the methods have similar correlation levels for Mary River and Lockyer Valley. On the other hand, ...
Specifying a correlation matrix is challenging in Fu,Liya,Wang,... - 《Journal of Multivariate Analysis An International Journal》 被引量: 1发表: 2016年 A two-part finite mixture quantile regression model for semi-continuous longitudinal data This paper develops a two-part finite mixture quantile...
effects across the size distribution of shocks. From the perspective of prudently supervising the financial system, network interconnectedness under extreme market conditions makes more sense than conventional average-based correlations (Betz et al.2016). Therefore, “extreme risk connectedness” has inevi...
Quantile regression models have become a widely used statistical tool in genetics and in the omics fields because they can provide a rich description of th