Seminar on Kernel density estimation with imputed group membership information

Events

  • 07 November 2012
  • Department of Mathematical Sciences, University of Copenhagen
  • Organiser: University of Copenhagen, Denmark
  • Contact: Helle Sørensen
  • E-mail: helle@math.ku.dk
  • Event Details

Speaker: Søren Feodor Nielsen, Copenhagen Business School, Center for Statistics

Abstract: 'The problem I will be considering in this talk is the estimation of an unknown density of a subgroup of continuous measurements when the group membership is missing for some of the observations. So, the data consists of a continuous outcome variable and a partially observed grouping variable and we are trying to estimate the conditional density of the outcome given the group.'

'If the probability of not having information on the group membership depends on the variable Y, then an estimator based on complete cases, i.e. restricting attention to observations with known group membership, will be biased. The traditional solution to this problem is to use imputation: "Guess" which group the observations belong to and base the estimator on the observations that are known or guessed to belong in the relevant group. Over the last two decades inverse probability weighting estimators have become increasingly popular in non- and semi-parametric missing data problems such as this one. Whereas imputation seems to be somewhat ad hoc, weighting is much more principled. It is however inefficient and imputation may be able to utilize more of the information in the data. In this talk I will compare weighting and different imputation strategies.'

Related Topics

Related Publications

Related Content

Site Footer

Address:

This website is provided by John Wiley & Sons Limited, The Atrium, Southern Gate, Chichester, West Sussex PO19 8SQ (Company No: 00641132, VAT No: 376766987)

Published features on StatisticsViews.com are checked for statistical accuracy by a panel from the European Network for Business and Industrial Statistics (ENBIS)   to whom Wiley and StatisticsViews.com express their gratitude. This panel are: Ron Kenett, David Steinberg, Shirley Coleman, Irena Ograjenšek, Fabrizio Ruggeri, Rainer Göb, Philippe Castagliola, Xavier Tort-Martorell, Bart De Ketelaere, Antonio Pievatolo, Martina Vandebroek, Lance Mitchell, Gilbert Saporta, Helmut Waldl and Stelios Psarakis.