Using The Biorthogonal Wavelet to Address the Problem Heterogeneity of Variance

Abstract

Thirty-two experimental units were inductive in this study based on a totally randomized design to provide impartial distribution of the treatments. The following analysis discussed the problem of heterogeneity in the variances by means of using biorthogonal wavelets together with the discrete wavelet transform (DWT), which gave a more robust model of the underlying data structure. All the moments that fade off the orthogonal wavelet family have been used in addition to the universal threshold (UT) value and the hard rule of thresholding. Comparative evaluations were performed across multiple performance criteria. The results revealed that the biorthogonal wavelet transform exhibited superior performance in stabilizing variance and improving analytical accuracy, demonstrating its potential as an effective tool for variance heterogeneity correction in experimental data analysis.