Heavy Lifting - thoughts and web finds by an economist
|I also contribute to Division of Labour||Load HL's Front Page|
Wednesday, May 26, 2010
Economists have come to rely upon government provided statistics of macroeconomic variables such as GDP, unemployment, and so forth. To the extent that these variables are mis-measured they introduce noise into our empirical analysis and can potentially lead to mistaken inference. In many cases this is not terribly important, but in many other areas it can be very important. Why?
To the extent that economic analysis informs public policy, the agnostic economist hopes that the data and empirical methodology employed are appropriate so that the truth of the situation can be uncovered.
Now it appears there is the increasing possibility that the unemployment numbers over the past several months are being manipulated, perhaps for political purposes at the top or for funding purposes at the Census Bureau. If these misleading employment figures are not systematically related with other variables, then the measurement error only influences the precision of empirical results. If, on the other hand, the measurement error is systematically related to other variables (say the impression of how the November mid-term elections might go) then this introduces potential bias into empirical results.
A word to the wise, in the future we might have to have a dummy variable that takes a value of one during the Obama administration years when using nationally provided employment data. That is a shame.
Comments: Post a Comment
Le Chai - galerie du vin
Posts that contain Craig Depken per day for the last 90 days.
Heavy Lifting's Main Page
Money I Found Today
Heavy Lifting - Firehose style (56k warning)
- Unemployment data might be wrong?
Modified maystar design
powered by blogger