The following blog has been published first at the Oxford University Politics Blog. Here is just the first part explaining the logic behind our BAFS method. Is it possible to have a more accurate prediction by asking people how confident they are that their preferred choice will win? As the Brexit referendum date approaches, the […]
Oraclum Intelligence Systems Ltd is a non-partisan start-up interested in experimental testing of forecasting models on real-life electoral data. We aim to use a Facebook survey of UK voters, along with our unique set of Bayesian forecasting methods to try and pick out the best and most precise prediction method in lieu of the upcoming […]
Our prediction method rests primarily upon our Facebook survey, where we use a variety of Bayesian updating methodologies to filter out internal biases in order to offer the most precise prediction.
Over the next month and a half, Oraclum I.S. Ltd. will be engaged in forecasting the outcome of the United Kingdom European Union membership referendum, colloquially referred to as the ‘Brexit’ referendum.
In our election forecasting model we combined election polls, previous electoral results and trends, and a number of socio-economic parameters that we uncovered to be important determinants of voter preferences. We used data on local level unemployment, exposure of the community to the 1991-1995 war for independence, and the educational structure of voters in each electoral district (these three factors carry the greatest weight in predicting voting patterns of domestic voters). In today’s text we will briefly show how each of the stipulated socio-economic variables is correlated with previous electoral results, and how they could serve as a partial predictor of future votes for a given party.
The most surprising result of the 2015 general election in Croatia was the success of the recently local party MOST, that appeared to have broken the duopoly of the two establishment center-left and center-right parties, the SDP and HDZ. MOST was able to attract a large majority of swing voters for themselves.
There is an almost tautological view amongst Croatian political analysts, the general public, the media, and the politicians themselves, that higher turnout benefits the left political parties, while lower turnout implies a much better electoral result for the right parties, which allegedly has a more disciplined voter base. This myth was most likely garnered after an impressive electoral victory of the left coalition in the 2000 elections, when turnout was significantly above average, but our analysis of the election results after 2000 shows that turnout does not in general affect electoral results, and that it has a minor influence only in districts which are party strongholds. In such districts the dominant party which mobilizes its support can achieve a significant electoral advantage over its opponent, which however does not necessarily have to translate into a national electoral victory.
As a part of an election coverage project initiated by the biggest daily newspaper in Croatia, Jutarnji list, we were given the opportunity to introduce, for the first time in this part of Europe, a prediction model of general elections which simultaneously uses election polls, previous election results, and a range of socio-economic data for a given electoral district.