We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:


Current browse context:


Change to browse by:

References & Citations


(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Statistics > Methodology

Title: Distribution-Free Predictive Inference For Regression

Abstract: We develop a general framework for distribution-free predictive inference in regression, using conformal inference. The proposed methodology allows construction of prediction bands for the response variable using any estimator of the regression function. The resulting prediction band preserves the consistency properties of the original estimator under standard assumptions, while guaranteeing finite sample marginal coverage even when the assumptions do not hold. We analyze and compare, both empirically and theoretically, two major variants of our conformal procedure: the full conformal inference and split conformal inference, along with a related jackknife method. These methods offer different tradeoffs between statistical accuracy (length of resulting prediction intervals) and computational efficiency. As extensions, we develop a method for constructing valid in-sample prediction intervals called rank-one-out conformal inference, which has essentially the same computational efficiency as split conformal inference. We also describe an extension of our procedures for producing prediction bands with varying local width, in order to adapt to heteroskedascity in the data distribution. Lastly, we propose a model-free notion of variable importance, called leave-one-covariate-out or LOCO inference. Accompanying our paper is an R package conformalInference that implements all of the proposals we have introduced. In the spirit of reproducibility, all empirical results in this paper can be easily (re)generated using this package.
Comments: 44 pages, 5 figures, 3 tables
Subjects: Methodology (stat.ME); Statistics Theory (math.ST); Machine Learning (stat.ML)
Cite as: arXiv:1604.04173 [stat.ME]
  (or arXiv:1604.04173v1 [stat.ME] for this version)

Submission history

From: Jing Lei [view email]
[v1] Thu, 14 Apr 2016 14:46:16 GMT (454kb,D)
[v2] Wed, 8 Mar 2017 18:52:14 GMT (679kb,D)

Link back to: arXiv, form interface, contact.