**Nonparametric predictive distributions based on conformal prediction.** / Vovk, Vladimir; Shen, Jieli; Manokhin, Valery; Xie, Minge.

Research output: Chapter in Book/Report/Conference proceeding › Chapter

Published

**Nonparametric predictive distributions based on conformal prediction.** / Vovk, Vladimir; Shen, Jieli; Manokhin, Valery; Xie, Minge.

Research output: Chapter in Book/Report/Conference proceeding › Chapter

Vovk, V, Shen, J, Manokhin, V & Xie, M 2017, Nonparametric predictive distributions based on conformal prediction. in A Gammerman, V Vovk, Z Luo & H Papadopoulos (eds), *Proceedings of Machine Learning Research: Proceedings of COPA 2016 (Sixth Symposium on Conformal and Probabilistic Prediction and Applications).* vol. 60, pp. 82-102.

Vovk, V., Shen, J., Manokhin, V., & Xie, M. (2017). Nonparametric predictive distributions based on conformal prediction. In A. Gammerman, V. Vovk, Z. Luo, & H. Papadopoulos (Eds.), *Proceedings of Machine Learning Research: Proceedings of COPA 2016 (Sixth Symposium on Conformal and Probabilistic Prediction and Applications) *(Vol. 60, pp. 82-102)

Vovk V, Shen J, Manokhin V, Xie M. Nonparametric predictive distributions based on conformal prediction. In Gammerman A, Vovk V, Luo Z, Papadopoulos H, editors, Proceedings of Machine Learning Research: Proceedings of COPA 2016 (Sixth Symposium on Conformal and Probabilistic Prediction and Applications). Vol. 60. 2017. p. 82-102

@inbook{65a68b2ba61340c4b8f03591244f80e9,

title = "Nonparametric predictive distributions based on conformal prediction",

abstract = "This paper applies conformal prediction to derive predictive distributions that are valid under a nonparametric assumption. Namely, we introduce and explore predictive distribution functions that always satisfy a natural property of validity in terms of guaranteed coverage for IID observations. The focus is on a prediction algorithm that we call the Least Squares Prediction Machine (LSPM). The LSPM generalizes the classical Dempster-Hill predictive distributions to regression problems. If the standard parametric assumptions for Least Squares linear regression hold, the LSPM is as efficient as the Dempster-Hill procedure, in a natural sense. And if those parametric assumptions fail, the LSPM is still valid, provided the observations are IID.",

keywords = "Conformal prediction., Least Squares, predictive distributions, regression",

author = "Vladimir Vovk and Jieli Shen and Valery Manokhin and Minge Xie",

year = "2017",

language = "English",

volume = "60",

pages = "82--102",

editor = "Alex Gammerman and Vladimir Vovk and Zhiyuan Luo and Harris Papadopoulos",

booktitle = "Proceedings of Machine Learning Research",

}

TY - CHAP

T1 - Nonparametric predictive distributions based on conformal prediction

AU - Vovk, Vladimir

AU - Shen, Jieli

AU - Manokhin, Valery

AU - Xie, Minge

PY - 2017

Y1 - 2017

N2 - This paper applies conformal prediction to derive predictive distributions that are valid under a nonparametric assumption. Namely, we introduce and explore predictive distribution functions that always satisfy a natural property of validity in terms of guaranteed coverage for IID observations. The focus is on a prediction algorithm that we call the Least Squares Prediction Machine (LSPM). The LSPM generalizes the classical Dempster-Hill predictive distributions to regression problems. If the standard parametric assumptions for Least Squares linear regression hold, the LSPM is as efficient as the Dempster-Hill procedure, in a natural sense. And if those parametric assumptions fail, the LSPM is still valid, provided the observations are IID.

AB - This paper applies conformal prediction to derive predictive distributions that are valid under a nonparametric assumption. Namely, we introduce and explore predictive distribution functions that always satisfy a natural property of validity in terms of guaranteed coverage for IID observations. The focus is on a prediction algorithm that we call the Least Squares Prediction Machine (LSPM). The LSPM generalizes the classical Dempster-Hill predictive distributions to regression problems. If the standard parametric assumptions for Least Squares linear regression hold, the LSPM is as efficient as the Dempster-Hill procedure, in a natural sense. And if those parametric assumptions fail, the LSPM is still valid, provided the observations are IID.

KW - Conformal prediction.

KW - Least Squares

KW - predictive distributions

KW - regression

M3 - Chapter

VL - 60

SP - 82

EP - 102

BT - Proceedings of Machine Learning Research

A2 - Gammerman, Alex

A2 - Vovk, Vladimir

A2 - Luo, Zhiyuan

A2 - Papadopoulos, Harris

ER -