Online Program Home
My Program

Abstract Details

Activity Number: 543 - Making Sense of Complex Featured Data with Statistical Methods
Type: Invited
Date/Time: Wednesday, July 31, 2019 : 2:00 PM to 3:50 PM
Sponsor: SSC
Abstract #300585
Title: Prediction for Federal Election by Joint Statistical Modeling
Author(s): Joan Fraser Hu* and Xin Shane Liu and Emma Qi Wen
Companies: Simon Fraser University and Shanghai University of Finance and Economics and Simon Fraser University
Keywords: Aggregated information; Combined data sources; Composite likelihood estimation; Multi-level data structure; Spatio-temporal correlation
Abstract:

Various efforts have been made to predict federal election outcomes. This talk begins with a review of the discrepancies between the real outcomes of recent Canadian federal elections and the predictions by the existing approaches such as the ones proposed by Grenier (www.threehundredeight.com) and Rosenthal (2011). We then critique the existing procedures to motivate an multi-level predictive model and its associated election prediction. We fit (train) the model with combination of the historical election outcomes with data from the Canadian Community Health Survey (CCHS) by 2011. Strategies are proposed for dealing with aggregated information, biased sampling, and spatio-temporal correlation. The fitted model is validated by the combined data after 2011 till 2015, the last Canadian federal election.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2019 program