Normalized nonconformity measures for automated valuation models

作者：

Highlights：

• Reliable conformal predictors are built for automated property price valuations.

• Nonconformity measures are constructed with data characteristics in mind.

• Model averaging over different models additionally improves predictive efficiency.

• The conformal predictors provide efficient prediction intervals of property prices.

摘要

•Reliable conformal predictors are built for automated property price valuations.•Nonconformity measures are constructed with data characteristics in mind.•Model averaging over different models additionally improves predictive efficiency.•The conformal predictors provide efficient prediction intervals of property prices.

论文关键词：Automated valuation model,Conformal predictor,Prediction interval

论文评审过程：Received 12 April 2020, Revised 26 August 2020, Accepted 4 May 2021, Available online 7 May 2021, Version of Record 10 May 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2021.115165