Improving Predictive Power of Conjoint Analysis by Constrained Parameter Estimation
研究发现，在联合分析的全轮廓法中，基于先验知识对参数施加约束能显著提升首选偏好的预测能力，但对权衡法的预测效度影响不大。
The prediction of first choice preferences by the full-profile method of conjoint analysis can be improved significantly by imposing constraints on parameters based on a priori knowledge of the ordering of part worths for different levels of an attribute. Constrained estimation however, has little effect on the predictive validity of the tradeoff method because the preference judgments within rows (or columns) of tradeoff tables have largely the same role as the constraints.