Ezra Klein from the Washington post created a forecasting model for the upcoming Presidential election.
The final model uses just three pieces of information that have been found to be particularly predictive: economic growth in the year of the election, as measured by the change in gross domestic product during the first three quarters; the president’s approval rating in June; and whether one of the candidates is the incumbent.
That may seem a bit thin. But it calls 12 of the past 16 elections right. The average error in its prediction of the two-party vote share is less than three percentage points.
It is interesting that there are only three parameters in the model. I should note that highly qualified academic experts informed this model, so this model is defensible (maybe I give my fellow academics too much credit (-: ).
Contrast these three parameters with Nate Silver’s model (from the NY Times): Presidential approval ratings, GDP growth rate in the fourth year of the incumbent’s term, and the ideology score of the challenger.
The Keys to the White House model by Allan Lichtman and Vladimir Keilis-Borok have 13 parameters that are all equally weighted. I wouldn’t necessarily conclude that more parameters means better accuracy. The equal weights may be a big limitation, since one could argue that not all of the 13 parameters carry the same weight in the mind of the voters. This may necessitate the use of additional parameters to accurately predict the race.
In summary, let’s look at a table of the parameters used in three models to predict the Presidential election. It’s amazing how little these models have in common and they are all “good” (i.e., they have predicted past election outcomes well). The one overlapping piece is the short-term change in the economy: if the GDP has improved in the recent past (i.e., this year!), then the President will be re-elected. This gives credence to the catch-phrase, “It’s the economy, stupid.”
|Keys to the White House (equally weights)||Nate Silver (unequal weights)||Ezra Klein (unequal weights)|
|Party Mandate: After the midterm elections, the incumbent party holds more seats in the US House of Representatives than after the previous midterm elections||-||-|
|Contest: There is no serious contest for the incumbent party nomination||-||-|
|Incumbency: The incumbent party candidate is the sitting president||-||Incumbency|
|Third party: There is no significant third party or independent campaign||-||-|
|Short term economy: The economy is not in recession during the election campaign||Economic Growth: G.D.P. growth during the election year itself||Economic Growth: change in gross domestic product during the first three quarters of the election year|
|Long term economy: Real per capita economic growth during the term equals or exceeds mean growth during the previous two terms||-||-|
|Policy change: The incumbent administration effects major changes in national policy||-||-|
|Social unrest: There is no sustained social unrest during the term||-||-|
|Scandal: The incumbent administration is untainted by major scandal||-||-|
|Foreign/military failure: The incumbent administration suffers no major failure in foreign or military affairs||-||-|
|Foreign/military success: The incumbent administration achieves a major success in foreign or military affairs||-||-|
|Incumbent charisma: The incumbent party candidate is charismatic or a national hero||-||-|
|Challenger charisma: The challenging party candidate is not charismatic or a national hero||-||-|
|-||Presidential approval ratings||Presidential approval ratings|
Nate Silver discusses the limitation of his model (or any model really) in forecasting the Presidential election that is worth posting here:
By design, [my forecast] is not an exceptionally precise forecast. There are all types of factors that the model does not explicitly consider, among them the possibility of third-party candidates or differences between the popular vote and the Electoral College. Moreover, voter perceptions about the economy, or the ideological positioning of the candidates, may differ in practice from what the objective data say about them. Rather than pretending to have all the answers, the model knows how much it doesn’t know and allows for a reasonably wide range of possible outcomes.