A novel risk-averse newsvendor problem is formulated as a chance-constrained problem to maximize profits from the sale of different products. In the newsvendor problem, both of the order quantity and the selling price are optimized for each product. A demand forecasting method based on Bayesian linear regression is employed to predict the probability distribution of demand for products from the selling prices. In order to solve the newsvendor problem, a revised artificial bee colony algorithm is proposed. The results of numerical experiments show that profits can be increased by optimizing the selling prices for products.