Inverse Matrix Problem in Regression for High-Dimensional Data Sets

Forhigh-dimensional chemometric data, the inverse matrix X t − 1 problem in regression models is a difficulty. Multicollinearity and identification result from problem. The usage of least absolute shrinkage selection operator (LASSO) partial squares are two existing ways dealing with (PLS). For regressing data sets, we used extended beta cube regression. proposed methods compared over near-infrared spectra biscuit dough Raman analysis contents polyunsaturated fatty acids (PUFA). reliable estimation, Monte Carlo cross-validation has been used. outperform based on root mean square error, indicating that can be for diverse high-dimensional sets.