Nonlinear expectation, including sublinear expectation as its special case, is a new and original framework of probability theory and has potential applications in some scientific fields, especially in finance risk measure and management. Under the nonlinear expectation framework, however, the related statistical models and statistical inferences have not yet been well established. The goal of this paper is to construct the sublinear expectation regression and investigate its statistical inference. First, a sublinear expectation linear regression is defined and its identifiability is given. Then, based on the representation theorem of sublinear expectation and the newly defined model, several parameter estimations and model predictions are suggested, the asymptotic normality of estimations and the mini-max property of predictions are obtained. Furthermore, new methods are developed to realize variable selection for high-dimensional model. Finally, simulation studies and a real-life example are carried out to illustrate the new models and methodologies. All notions and methodologies developed are essentially different from classical ones and can be thought of as a foundation for general nonlinear expectation statistics.