Introduction to Statistical Inference and Models

2024-10-23

Statistical inference

  • Statistical Inference is the branch of statistics dedicated to distinguishing patterns arising from signal versus those arising from chance.

  • It is a broad topic and we review the basics using polls as a motivating example.

  • We motivate the concepts with election forecasting as a case study.

Statistical inference

  • The day before the 2008 presidential election, Nate Silver’s FiveThirtyEight stated that “Barack Obama appears poised for a decisive electoral victory”.

  • They went further and predicted that Obama would win the election with 349 electoral votes to 189, and the popular vote by a margin of 6.1%.

  • FiveThirtyEight also attached a probabilistic statement to their prediction claiming that Obama had a 91% chance of winning the election.

Statistical inference

  • The predictions were quite accurate since, in the final results, Obama won the electoral college 365 to 173 and the popular vote by a 7.2% difference.

  • Their performance in the 2008 election brought FiveThirtyEight to the attention of political pundits and TV personalities.

  • Four years later, the week before the 2012 presidential election, FiveThirtyEight’s Nate Silver was giving Obama a 90% chance of winning despite many of the experts thinking the final results would be closer.

Statistical inference

  • Political commentator Joe Scarborough said during his show

Anybody that thinks that this race is anything but a toss-up right now is such an ideologue they’re jokes.

Statistical inference

  • To which Nate Silver responded via Twitter:

If you think it’s a toss-up, let’s bet. If Obama wins, you donate $1,000 to the American Red Cross. If Romney wins, I do. Deal?

Statistical inference

  • In 2016, Silver was not as certain and gave Hillary Clinton only a 71% of winning.

  • In contrast, many other forecasters were almost certain she would win.

  • She lost.

Statistical inference

  • But 71% is still more than 50%, so was Mr. Silver wrong?

  • What does probability mean in this context anyway?

  • We will demonstrate how the probability concepts covered in the previous part can be applied to develop statistical approaches that render polls effective tools.

Statistical inference

  • Forecasting an election is a more complex process that involves combining results from 50 states and DC.

  • We will learn the statistical concepts necessary to define estimates and margins of errors for the popular vote, and show how these are used to construct confidence intervals.

  • Once we grasp these ideas, we will be able to understand statistical power and p-values, concepts that are ubiquitous in the academic literature.

Statistical inference

  • We will then aggregate data from different pollsters to highlight the shortcomings of the models used by traditional pollsters and present a method for improving these models.

  • To understand probabilistic statements about the chances of a candidate winning, we will introduce Bayesian modeling.

  • Finally, we put it all together using hierarchical models to recreate the simplified version of the FiveThirtyEight model and apply it to the 2016 election.