英文文献：Networking Your Way to a Better Prediction: Effectively Modeling Contingent Valuation Survey Data-通过网络实现更好的预测:有效地建模条件估值调查数据
英文文献作者：Bergtold, Jason S.,Taylor, Daniel B.,Bosch, Darrell J.
英文文献摘要：
The purpose of this paper is to empirically compare the out-of-sample predictive capabilities of artificial neural networks, logit and probit models using dichotmous choice contingent valuation survey data. The authors find that feed-forward backpropagation artificial neural networks perform relatively better than the binary logit and probit models with linear index functions. In addition, guidelines for modeling contingent valuation survey data and how to estimate median WTP using artificial neural networks are provided.

本文的目的是利用海量选择条件估值调查数据，对人工神经网络、logit和probit模型的样本外预测能力进行实证比较。研究发现，前馈反向传播人工神经网络的性能要优于具有线性索引函数的二值logit和probit模型。此外，还提供了建模条件估值调查数据和如何使用人工神经网络估计期望值中值的指导方针。