Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Poll of polls: A compositional loess model

Bergman, Jakob LU orcid and Holmquist, Björn LU orcid (2014) In Scandinavian Journal of Statistics 41(2). p.301-310
Abstract
A large number of polls on party preferences are published today. In order to get an estimate of the changes in political opinion, the polls may be combined into a poll of polls. We discuss a method for combining polls using the fact that they are compositions and respecting the properties of the compositional sample space (the simplex). The method is easily implemented and the estimate may be computed in linear time. We provide an example with Swedish data from year 2007 to 2010. The method also allows us to present the deviations between the estimated compositions and the observed. In the

data set, we note e.g. differences between different polling institutes.
Please use this url to cite or link to this publication:
author
and
organization
publishing date
type
Contribution to journal
publication status
published
subject
keywords
party preferences, compositional weighted least sum of squares, simplex, smoothing
in
Scandinavian Journal of Statistics
volume
41
issue
2
pages
301 - 310
publisher
Wiley-Blackwell
external identifiers
  • wos:000335388400003
  • scopus:84899981634
ISSN
1467-9469
DOI
10.1111/sjos.12023
language
English
LU publication?
yes
id
b43d7a29-b372-43c3-bfa7-3e72bc68b477 (old id 3437710)
alternative location
http://onlinelibrary.wiley.com/doi/10.1111/sjos.12023/abstract
date added to LUP
2016-04-01 11:01:02
date last changed
2022-03-27 21:43:58
@article{b43d7a29-b372-43c3-bfa7-3e72bc68b477,
  abstract     = {{A large number of polls on party preferences are published today. In order to get an estimate of the changes in political opinion, the polls may be combined into a poll of polls. We discuss a method for combining polls using the fact that they are compositions and respecting the properties of the compositional sample space (the simplex). The method is easily implemented and the estimate may be computed in linear time. We provide an example with Swedish data from year 2007 to 2010. The method also allows us to present the deviations between the estimated compositions and the observed. In the<br/><br>
data set, we note e.g. differences between different polling institutes.}},
  author       = {{Bergman, Jakob and Holmquist, Björn}},
  issn         = {{1467-9469}},
  keywords     = {{party preferences; compositional weighted least sum of squares; simplex; smoothing}},
  language     = {{eng}},
  number       = {{2}},
  pages        = {{301--310}},
  publisher    = {{Wiley-Blackwell}},
  series       = {{Scandinavian Journal of Statistics}},
  title        = {{Poll of polls: A compositional loess model}},
  url          = {{https://lup.lub.lu.se/search/files/51894216/Bergman_Holmquist_SJS_final.pdf}},
  doi          = {{10.1111/sjos.12023}},
  volume       = {{41}},
  year         = {{2014}},
}