Greben, JM2008-12-122008-12-122006-02-28Greben, JM. 2006. Theory of quantitative trend analysis and its application to the South African elections. CSIR Research and Innovation Conference: 1st CSIR Biennial Conference, CSIR International Convention Centre Pretoria, 27-28 February 2006, pp 15http://hdl.handle.net/10204/2732CSIR Research and Innovation Conference: 1st CSIR Biennial Conference, CSIR International Convention Centre Pretoria, 27-28 February 2006In this paper the author discusses a quantitative theory of trend analysis. Often trends are based on qualitative considerations and subjective assumptions. In the current approach the author makes use of extensive data bases to optimise the so-called trend matrix. The theory is applied to the South African elections. The record of election results in allvoting districts for subsequent elections can be used to give a reliable estimate of the election trends. These results can then be compared to less data-intensive methods, which partially rely on common sense assumptions and on readily available subset(s) of election results. We analyse the validity and appropriateness of these simplified assumptions. We also discuss how these techniques can be used in other circumstances, for example in the field of marketingenQuantitative trend analysisElections trendsTrend matrixElection resultsSouth AfricaConference PresentationGreben, J. (2006). Theory of quantitative trend analysis and its application to the South African elections. http://hdl.handle.net/10204/2732Greben, JM. "Theory of quantitative trend analysis and its application to the South African elections." (2006): http://hdl.handle.net/10204/2732Greben J, Theory of quantitative trend analysis and its application to the South African elections; 2006. http://hdl.handle.net/10204/2732 .TY - Conference Presentation AU - Greben, JM AB - In this paper the author discusses a quantitative theory of trend analysis. Often trends are based on qualitative considerations and subjective assumptions. In the current approach the author makes use of extensive data bases to optimise the so-called trend matrix. The theory is applied to the South African elections. The record of election results in allvoting districts for subsequent elections can be used to give a reliable estimate of the election trends. These results can then be compared to less data-intensive methods, which partially rely on common sense assumptions and on readily available subset(s) of election results. We analyse the validity and appropriateness of these simplified assumptions. We also discuss how these techniques can be used in other circumstances, for example in the field of marketing DA - 2006-02-28 DB - ResearchSpace DP - CSIR KW - Quantitative trend analysis KW - Elections trends KW - Trend matrix KW - Election results KW - South Africa LK - https://researchspace.csir.co.za PY - 2006 T1 - Theory of quantitative trend analysis and its application to the South African elections TI - Theory of quantitative trend analysis and its application to the South African elections UR - http://hdl.handle.net/10204/2732 ER -