|
Early Predictions of U.S. Presidential Election Winners |
|
Written by Administrator
|
|
Wednesday, 06 May 2009 12:58 |
At his Keynote talk at the International Symposium on Forecasting (ISF)
in Hong Kong this June, Scott Armstrong will discuss a long-term
forecasting project. As part of this, he would like to obtain prior
expectations, and to present these at the ISF. Read more.
|
|
Written by Andreas Graefe
|
|
Tuesday, 18 November 2008 09:43 |
|
On his homepage, Ray Fair analyzes the performance of the presidential vote-equation. He also links to a new paper, forthcoming in the American Journal of Political Science. |
|
Read more...
|
|
New Scientist covers Polly |
|
Written by Andreas Graefe
|
|
Tuesday, 11 November 2008 22:09 |
In an article on election forecasting, published before the election, the New Scientist covered PollyVote.
|
|
Written by Andreas Graefe
|
|
Thursday, 06 November 2008 14:02 |
Polly provides the graphs and data for this years election. Take a look and compare how Polly performed compared to her components as well as to raw polls and the original forecasts of the Iowa Electronic Markets.
|
|
Written by Andreas Graefe
|
|
Wednesday, 05 November 2008 15:29 |
Polly revealed her data. Click here to access the data file containing the forecasts derived from the RCP poll average, the IEM, the expert panel as well as the quantitative models. For now, we entered the current election results from Wednesday afternoon, indicating that John McCain received 46.9% of the popular two-party-vote. You can change this value in cell M2 of the PollyVote main sheet. Click here for the PollyElectoral Vote data.
|
|
Ray Fair's model performance |
|
Written by Andreas Graefe
|
|
Wednesday, 05 November 2008 11:34 |
|
Ray Fair's post-mortem on his presidential vote equation shows that over a period of two years it performed very well in forecasting yesterday's outcome. Read more. |
|
Written by Andreas Graefe
|
|
Wednesday, 05 November 2008 08:27 |
|
For the second time in a row, Polly called a presidential election with a perfect or near-perfect forecast. In 2004, the morning after the voters had had their say it appeared that Polly had hit the bull's eye, at 51.5% for President Bush. When the final official results were in, however, it turned out that that was not quite right. The Pollyvote was off-by a mere one-fourth of one percent! This morning preliminary totals show the Obama-Biden ticket taking 53% of the two-party vote, exactly what Polly predicted on Election Eve. Once again, Polly has shown that combining election forecasts within and across methods, with polls, quantitative models, a prediction market and experts' forecasts averaged and weighted equally, is hard to beat. |
|
Polly's Final Forecast: Obama will win |
|
Written by Andreas Graefe
|
|
Tuesday, 04 November 2008 11:18 |
|
For more than a year, well before the candidates of either party were known, Polly has been forecasting a defeat of the incumbents and hence a political party turnover in the White House. Today, Polly issues her final forecast: The Democratic ticket of Barack Obama and Joseph Biden will win the presidency with 53.0% of the two-party vote and 354 votes in the Electoral College (vs. McCain and Palin's 47.0% and 184 votes, respectively). |
|
Polly's final expert forecast |
|
Written by Andreas Graefe
|
|
Monday, 03 November 2008 08:31 |
|
Polly's elections experts have now weighed in with their final forecast. In a October 31 - November 2 survey, the median prediction among 13 panelists was that Republican John McCain will receive 47.5% of the major-party vote -- and thus that Democrat Barack Obama will garner 52.5%. Thus, the experts' forecast remained stable since the last round in mid-October. |
|
Read more...
|
|
Predicting Elections from Demographic Information about Candidates |
|
Written by Andreas Graefe
|
|
Monday, 03 November 2008 07:50 |
|
In one other effort to increase our knowledge about who is going to win the election, Armstrong and Graefe are working on an approach on predicting the U.S. presidential election based on the demographic information about candidates. Similar to PollyIssues, they again use the index method. Their work is in very preliminary status but first results are encouraging. |
|
Read more...
|
|
Written by Andreas Graefe
|
|
Saturday, 01 November 2008 17:45 |
|
For the last time before the election, Polly reached out to her expert panel. We are excited to hear what they will tell us. Stay tuned! |
|
Updated forecast from the "Bread and Peace" model |
|
Written by Andreas Graefe
|
|
Friday, 31 October 2008 10:11 |
|
Douglas Hibbs provided an updated forecast for the "Bread and Peace" model. His new forecast is 46.25% for McCain. |
|
Read more...
|
|
Ray Fair forecast updated |
|
Written by Andreas Graefe
|
|
Friday, 31 October 2008 09:33 |
|
Ray Fair provided his final forecast, predicting Obama to win, with a vote-share of 51.9% vs. McCain 48.1%. |
|
Read more...
|
|
Written by Andreas Graefe
|
|
Wednesday, 29 October 2008 12:01 |
|
Graefe and Armstrong updated PollyIssues, including recent polls until October 23. Their forecast remained unchanged, still predicting Obama to win with a vote-share of 52.5%. |
|
Written by Andreas Graefe
|
|
Tuesday, 28 October 2008 09:25 |
|
Jay DeSart and Tom Holbrook updated their model. Their latest forecast from October 25 now predicts Obama to receive 53.82% of the two-party vote. |
|
|