LAUSR

Regression is widely applied in many fields. Regardless of the types of regression, we often assume that the observations are precise. However, in real-life circumstances, this assumption can only be met sometimes, which means the traditional regression methods can result in significant imprecise or biased predictions. Consequently, uncertain regression models might provide more accurate and meaningful results under these circumstances. In this article, we provide the residual analysis of uncertain Gompertz regression model, as well as the corresponding forecast value and confidence interval. Finally, we give a numerical example of uncertain Gompertz regression model.