A new uncertain linear regression model based on equation deformation

When the observed data are imprecise, uncertain regression model is more suitable for linear analysis. Least squares estimation can fully consider given and minimize sum of residual error effectively solve equation imprecisely data. On basis uncertainty theory, this paper presents an deformation method solving unknown parameters in equations. We first establish one-dimensional then extend it to case multiple model. also combine with Cramer’s rule matrix propose elementary transformation equation. Numerical example show that