Information Technologies and Computer Engineering

The non-linear regression model to estimate the software size of VB-based information systems is constructed on the basis of normalization of the four-dimensional non-Gaussian data set (actual software size in the thousand lines of code, the total number of classes, the total number of relationships and the average number of attributes per class in conceptual data model from 32 systems) by the Johnson multivariate transformation for SB family. Comparison of the constructed model with the linear regression model and non-linear regression models based on the decimal logarithm and the Johnson univariate transformation is performed. Thе constructed model, in comparison with other regression models (both linear and non-linear), has a larger multiple coefficient of determination, a smaller value of the mean magnitude of relative error and smaller widths of the confidence and prediction intervals of non-linear regression

References