In: Statistics and Probability
How would you go about creating a linear regression model to predict the 2020 Presidential Election?
It may be hard to believe, but the 2020 presidential election is roughly many days away. In what is shaping up to be another contentious year in politics, forecasts predict an unprecedented number of voters will turn out to vote for nominees that will undoubtedly have very different agendas for our country’s future.
As the Democratic party seeks to win over ex-leader voters over the next year, I thought it would be interesting to take a look back at the 2016 election results to identify factors that led to his success. During my analysis, it became quite clear that certain values and ideologies were predictive in estimating the percent of voters that cast their vote for President on a state-wide level.
By using linear regression, I created a model that captured close to 87 percent of the variability in the proportion of votes for President .
Now, for those that are more statistically survey, this model would by no means be effective at predicting the outcome of the 2020 election (or even the 2016 election), but it does tell us some valuable things about the way voters feel about certain “hot-button” issues and the effect they have on their decision to vote for President or not.
Here we analyze the data or then we say or predict who would win or not ,Here there is major role of data analysis on the basis of data.