Re: Data Manipulation

[Doug Henwood <dhenwood@xxxxxxxxx>]

michael perelman quoted:

> Taking a page from the
> Bureau of Labor Statistics handbook (birth-death adjustment, anyone?),

That's stupid. The birth-death model works quite well, and much
better than the "bias adjustment" that preceded it. Recent benchmark
revisions to employment have been very small by historical standards
- and nothing like the big revisions of the early 1990s.

> Don't care much for that? Then consider what BLS hath wrought. Their GDP
> revisions for 2005's 1Q border on the absurd.

The BLS doesn't do GDP - the BEA does.

>  In order to crank GDP from
> its disappointing initial reading of 3.1% to the more vigorous final
> 3.8%, the BLS had to make some sketchy adjustments. Primary amongst
> their changes was (I am not making this up!) an actual decrease in home
> prices for Q1.

The writer forgot that growth estimates for 2001-2004 in the annual
benchmark revisions were knocked down from 3.1% to 2.8%.

This sort of critique is dumb. The BLS and BEA do a very good job.
It's hard work getting this stuff out in a month or three. Revisions
go in both directions - there's no pattern to them. Nothing is being
manipulated for political reasons.

Doug