In brand choice, consumers usually form a consideration set first and then choose products from the consideration sets. In this paper, the consumer’s choice behavior is modeled as a two-stage decision-making process: the consideration set formation stage and the product choice stage. In the first stage, we assume that consumers construct consideration sets based on the ranking of the overall utility of products. In the second stage, consumers choose a product with the minimum expected regret by comparing each attribute of products in the consideration set. In our consideration set model, the first stage is a multiattribute compensatory decision process, and the second stage is a noncompensatory choice process. Compared with a single-stage conditional logit model with utility maximization and a single-stage random regret minimization model, Monte Carlo simulation results show that our two-stage consideration set model is superior to the single-stage choice model based on both the mean bias and the root mean squared error of the estimator of model parameter.