(This article was originally published at Learning From Data » Statistics, and syndicated at StatsBlogs.)
Finally, the wait is over. Welcome back, regular ads! It is depressing (and exciting in a certain degree) to live in Columbus (OH) during an election year, especially as a statistician.
Watching political ads talking about numbers and statistics is depressing. These guys are really good at picking up a needle from a haystack and stick it to a particular group of voters.
On the other hand, it is exciting to see everyone trying to predict the outcome of the election using various methods (statistical or not):
- Survey methods: national polls, state polls, exit polls, endless telephone surveys.
- Model based methods: just list one the most famous FiveThrityEight by Nate Silver, the author of a new book The Signal and the Noise.
- Prediction based on wisdom of a crowd: intrade.com, where you can buy candidates as stocks
And of course, the last resource for anyone to use: Guesstimation. See what wikipedia says:
Guesstimate is an informal English contraction of guess and estimate, first used by American statisticians in 1934 or 1935. It is defined as an estimate made without using adequate or complete information,or, more strongly, as an estimate arrived at by guesswork or conjecture. [...]
Guesstimation techniques are used:
- in physics, where the use of guesstimation techniques to solve Fermi problems is taught as a useful skill to science students.
- in cosmology, where the Drake equation is a well-known guesstimation method.
- in economics, where economic forecasts and statistics are often based on guesstimates. (my comment: this is beyond being funny.)
Let the guesstimation start! Finally, it is close to be over. We will see the result soon, and hopefully.
Reporting from Ohio, ground zero for the election.
Please comment on the article here: Learning From Data » Statistics
