2006
DOI: 10.5784/22-1-35
|View full text |Cite
|
Sign up to set email alerts
|

A model for election night forecasting applied to the 2004 South African elections

Abstract: A novel model has been developed to predict elections on the basis of early results. The electorate is clustered according to their behaviour in previous elections. Early results in the new elections can then be translated into voter behaviour per cluster and extrapolated over the whole electorate. This procedure is of particular value in the South African elections which tend to be highly biased, as early results do not give a proper representation of the overall electorate. In this paper we explain the metho… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
8
0

Year Published

2011
2011
2018
2018

Publication Types

Select...
4
1

Relationship

0
5

Authors

Journals

citations
Cited by 6 publications
(8 citation statements)
references
References 11 publications
0
8
0
Order By: Relevance
“…As indicated in the previous section, this model was developed for the South African elections and has been applied during the 1999, 2004, 2009, 2014 general elections and the 2000, 2006, 2011, 2016 municipal elections. The mathematical details of this forecasting model can be obtained in (Greben et al, 2006).…”
Section: Methodsmentioning
confidence: 99%
See 1 more Smart Citation
“…As indicated in the previous section, this model was developed for the South African elections and has been applied during the 1999, 2004, 2009, 2014 general elections and the 2000, 2006, 2011, 2016 municipal elections. The mathematical details of this forecasting model can be obtained in (Greben et al, 2006).…”
Section: Methodsmentioning
confidence: 99%
“…This team represents a successful collaboration of computer scientists, GIS practitioners and statisticians who have developed a model to produce real-time updates to forecasts as the "live" information on vote counts is released. The method involves using clusters based on past voting patterns to predict final election outcomes during election nights, once a sample of between 7 to 10 percent of voting districts have been declared, and has been used successfully in a number of South African elections as illustrated in Greben et al (2005) and Greben et al (2006). After the success achieved in the 2016 South African elections, in which the forecasts correctly predicted final outcomes which were not generally expected in the run-up to the elections, the question arose as to whether the method could be generally applicable to elections carried out in other countries.…”
Section: Introductionmentioning
confidence: 99%
“…The foundation of election night forecasting lies in ecological regression models [2,5,6,7,4]. The performance of a party at a current election is considered to be a linear combination of the performances of all parties at a past election.…”
Section: Introductionmentioning
confidence: 99%
“…The aim of this paper is not to describe the model since a detailed outline of this model is given by Greben et al [12] as well as by Greben et al [13]. Further, Ittmann [22] gives some personal perspectives and experiences of interacting with the media in sharing the forecasts produced by this model.…”
Section: Model and Assumptionsmentioning
confidence: 99%
“…With appropriate samples sizes a forecast of the outcome is presented almost as soon as the election booths close. In South Africa a forecasting model was developed which endeavours to forecast the final outcome based on the first results as they become available [12,13]. The model is thus used after the event to forecast the final outcome.…”
Section: Literature Reviewmentioning
confidence: 99%