On expectile-assisted inverse regression estimation for sufficient dimension reduction

Projection expectile regression for sufficient dimension reduction

Many existing sufficient dimension reduction methods are designed for regression with predictors that are elliptically distributed, which limits their application in real data analyses. Projection expectile regression (PER) is proposed as a new linear sufficient dimension reduction method for handling complex predictor structures, which includes continuous, discrete, and mixed predictor variables. PER requires the link function between the response and the predictor to be monotone, but not necessarily smooth, which makes it suitable for handling stratified response surfaces. By design, PER does not involve matrix inversion or high-dimensional smoothing. Thus, PER is ideal for controlling problems associated with multicollinearity, high dimensionality, and sparsity in the predictor. An extensive simulation study demonstrates the performance of projection expectile regression in synthetic data. A real data analysis of health insurance charges in the United States is also provided. The asymptotic properties of the PER estimator are included as well.

Co3O4@TiO2@Y2O3 nanocomposites for a highly sensitive CO gas sensor and quantitative analysis

In order to predict the early failure of organic insulator, Co₃O₄@TiO₂@Y₂O₃ nanocomposites was prepared and characterized (XRD, SEM, EDS, FTIR, UV-vis-NIR, XPS) to detect decomposition CO gas. A simple experimental platform was built to verify the excellent adsorption, stability, selectivity and repeatability of the composite. Then, the mechanism of adsorption enhancement was analyzed by heterojunction. Aiming at 170 sets of gas sensing data sets, Successive Projections Algorithm (SPA) was used to extract data features, and grey wolf optimization vector machine regression (GWO-SVR) model was established to predict carbon monoxide concentration. The correlation coefficient (R_P), root mean square error (RMSEP) and calculation time of prediction set were 99.3025%, 0.0418 and 1.47 s, respectively. Therefore, the combination of the superior properties of a composite sensitive material and the small sample quantitative prediction model is a promising method for gas sensors in the future.