Drawing Blanks

Premature Optimization is a Prerequisite for Success

Russian Elections: the facts (?)

leave a comment »

Finally, all statisticians, mathematicians, sociologists, politologists and analysts of all sorts can take some rest. There is no need to mine the data or come up with models in order to detect and measure election fraud anymore.

Thanks to http://ruelect.com/en/

The folks are collecting copies of the tally sheets (“voting protocols”) obtained by observers on the day of voting and comparing them to the official results released by the Central Electoral Committee on the next day.

If their data is sound, the discrepancies are massive, stupid, and they are of course in favor of the ruling party.

I’m not going to analyze their data. I took a look at some samples. I still believe that the fraud did not alter the statistics of the elections in any detectable way. In the majority of samples that I looked at, the fraud did not result in too high vote ratio change, and the turnout ratio was almost never altered. All the tails are still there, they just became a bit fatter.

When I’m saying that I’m not going to study the data, I don’t mean that nobody should study it. Somebody must study it! For example, it would be really useful to see how much the forged data deviates from the social conformity curve (it does deviate!). That might help develop methodology to detect such fraud in the future.

I wonder why aren’t courts in Russia looking at this stuff.

Advertisements

Written by bbzippo

01/06/2012 at 5:48 am

Posted in Uncategorized

Tagged with

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: