"according
to
the
Bread
and
Peace
model
per
capita
real
income
growth
rates
must
average
out
at
nearly
6
percent
after
2012:q2
for
Obama
to
have
a
decent
chance
of
re-‐election."
You can get to the whole paper from here (and obviously a hat tip goes to Brendan).
Beyond this bombshell, the paper is well worth reading as Hibbs excoriates his competitors for using ad-hoc and ex-post dummy variables as well as endogenous approval ratings as explanatory variables in their vote-share equations.
Seems like he's conflating prediction and explanation. Who cares if approval is endogenous if it helps us predict. I've some of his competitors speak, and mostly they don't frame their work in terms of pinning down causal effects.
ReplyDeleteI took an Econometrics course back in 2004 where we read Fair's book and came up with our own models to predict presidential elections [we could choose national vote percentage or electoral college outcome]. I had three variables and disposable income growth for the preceding eight quarters was one the variables I used, too. It turns out that another statistically significant variable (that also explained a great deal with a substantial R-square) was a 28 year cycle where the parties' electoral votes from seven election cycles in the past explained a whole lot about the electoral college outcome 28 years into the future. 1984 was a pretty good year for Republicans, wasn't it?
ReplyDeleteMakes about as much sense as predicting hurricanes with straight lines.
ReplyDeleteUN PRESIDENT TIM KALEMKARIAN, US PRESIDENT TIM KALEMKARIAN, US SENATE TIM KALEMKARIAN, US HOUSE TIM KALEMKARIAN: BEST MAJOR CANDIDATE.
ReplyDeleteUN PRESIDENT HUGH HEWITT, US PRESIDENT HUGH HEWITT, US SENATE HUGH HEWITT, US HOUSE HUGH HEWITT, GOVERNOR HUGH HEWITT, LT GOVERNOR HUGH HEWITT: BEST MAJOR CANDIDATE.
ReplyDeleteUN PRESIDENT MIKE GALLAGHER, US PRESIDENT MIKE GALLAGHER, US SENATE MIKE GALLAGHER, US HOUSE MIKE GALLAGHER: BEST MAJOR CANDIDATE.
ReplyDelete