Semiparametric Exponential Families for Heavy-Tailed Data
Preprint
Abstract
We propose a semiparametric method for fitting the tail of a heavy-tailed population given a relatively small sample from that population and a larger sample from a related background population. We model the tail of the small sample as an exponential tilt of the better-observed large-sample tail, using a robust sufficient statistic motivated by extreme value theory. In particular, our method induces an estimator of the small-population mean, and we give theoretical and empirical evidence that this estimator outperforms methods that do not use the background sample. We demonstrate substantial efficiency gains over competing methods in simulation and on data from a large controlled experiment conducted by Facebook.Keywords
All Related Versions
This publication has 14 references indexed in Scilit:
- Likelihood-Based Procedures for Threshold Diagnostics and Uncertainty in Extreme Value ModellingJournal of the Royal Statistical Society Series B: Statistical Methodology, 2012
- Proportional likelihood ratio models for mean regressionBiometrika, 2012
- A note on profile likelihood for exponential tilt mixture modelsBiometrika, 2009
- Merging Information for Semiparametric Density EstimationJournal of the Royal Statistical Society Series B: Statistical Methodology, 2004
- Estimating the Mean of Heavy-Tailed DistributionsExtremes, 2003
- A Diagnostic for Selecting the Threshold in Extreme Value AnalysisJournal of the Royal Statistical Society Series B: Statistical Methodology, 2001
- Estimating the mean of a heavy tailed distributionStatistics & Probability Letters, 2001
- A Semiparametric Approach to the One-Way LayoutTechnometrics, 2001
- An Introduction to Statistical Modeling of Extreme Values2001
- On Some Simple Estimates of an Exponent of Regular VariationJournal of the Royal Statistical Society Series B: Statistical Methodology, 1982