To retrieve the lead content in the soil from hyperspectral remote sensing data, we should select bands or do some transformation first. In this paper, we compared Least Angle Regression, which is a modest forward choose method, and least squares regression and partial least squares regression based on genetic algorithm. As a validation result of  the test area in Jianning, Nanjing, regression results with band selection are better than those without. Although Least Angle Regression, partial least squares regression with genetic algorithm can reach 70% training correctness, the latter based on genetic algorithm is better, because it can reach a larger solution space. At last, we conclude that partial least squares regression is a good choice for the lead content retrieval in soil by hyperspectral remote sensing data, and genetic algorithm can improve the retrieval by band selection promisingly.