Munich Personal RePEc Archive

Forecasting 2016 US Presidential Elections Using Factor Analysis and Regression Model

Sinha, Pankaj and Srinivas, Sandeep and Paul, Anik and Chaudhari, Gunjan (2016): Forecasting 2016 US Presidential Elections Using Factor Analysis and Regression Model.


Download (894kB) | Preview


The paper categorizes factors responsible for forecasting the outcome of U.S. presidential election 2016 using factor analysis, which groups the various economic and non-economic parameters based on the correlation among them. The major economic factor significant in 2016 US presidential election is the growth of the economy, and the ‘anti-incumbency factor that signifies how long the incumbent party has been controlling the White House is found to be an important non-economic factor likely to play a dominant role in the election.

The dependent variables considered are the vote shares of the nominees of the incumbent and the non-incumbent majority party candidates. The forecast is calculated by running a regression of the significant factors, obtained through factor analysis technique, on the incumbent party vote share as well as on the non-incumbent party vote share.

The proposed models forecast the vote share of Democrat candidate Mrs. Hillary Clinton to be 45.59% with a standard error of ±2.32% and that of Republican candidate Mr. Donald Trump to be 39.51% with a standard error of ±3.87%. Hence, the models built in the paper signal a comfortable margin of victory for the Presidential nominee of the incumbent party, Hillary Clinton. The study re-establishes the notion that the non-economic factors have a greater influence on the outcomes of election as compared to the economic factors, as some of the important economic factors such as inflation and unemployment rate failed to establish their significance.

Atom RSS 1.0 RSS 2.0

Contact us: mpra@ub.uni-muenchen.de

This repository has been built using EPrints software.

MPRA is a RePEc service hosted by Logo of the University Library LMU Munich.