HN Explorer

9 comments

[ 3.4 ms ] story [ 51.3 ms ] thread
Volpe
I wonder if there is a way, to gather this info separately from the government and publish results? Or at least publish an accuracy of the public numbers...

Could be an interesting space.

wisty
It could also be a little bit illegal, and possibly considered espionage.
seanmcdirmid
Definitely would violate state secrets laws. The government wants a monopoly on this kind of information and will prevent anyone else from just going out and collecting statistics/polling on their own. Look at the fit they rose when the American embassy published pollution sastitics for what they measured (which is only legal because the embassy is sovereign territory...the real reason why they block twitter!).
ThomPete
One of the problems that are being solved now was that the actual data to determine growth was made by local officials. There have apparenttly been some speculation as to whether they have been a little to eager to prove them selves and inflated the numbers.
diminish
The public should keep a suspicious eye on every stat published; only that way, we would notice the Libor scandal, 2008 mortgage related valuations, or a lot of countries from Argentina to Turkey playing with their major economic stat reporting such as inflation, GDP, foreign trade etc.
reitzensteinm
In the US, the government just openly changes the methodology, which is forgotten shortly after. Seems to work just as well as falsifying the stats, with none of the risk.

http://en.wikipedia.org/wiki/Shadowstats.com

I'm not recommending Shadow Stats specifically - I have no idea how accurate or biased they are.

But there's a clear need for this information in order to do any kind of historical comparisons of things as basic as inflation and unemployment numbers.

EzGraphs
Statistical and data mining techniques have been used to detect fraud in elections and financial statements.

http://www.economist.com/node/8733747?story_id=8733747

https://en.wikipedia.org/wiki/Data_Analysis_Techniques_for_F...

The NY Times reported last month that

Studies by Goldman Sachs and other institutions over the years have strongly suggested that Chinese statisticians smooth out the quarterly growth figures, underreporting growth during boom years and overstating growth during economic downturns.

http://www.nytimes.com/2012/06/23/business/global/chinese-da...

So private corporations (who granted have their own agenda and biases) make assessments of economic numbers reported.

It would be surprising that there is not publicly available studies that report the general reliability of economic statistics reported by a country. It would even be interesting to reported statistics aggregated in a way that allows for easy comparison between numbers reported by different countries. Something like the Index of Economic freedom: http://www.heritage.org/index/default

(I am not an economist - so can't say I have done an extensive analysis of available sources...)