Bernard Condon gives another reason to be cautious about the unemployment figures that were released yesterday that I hadn’t thought of.
The numbers are seasonally adjusted. That means that a smoothing function has been applied to the numbers to eliminate temporary blips in hiring and firing. But that, in turn, raises a serious mathematical problem: when the amount of smoothing that’s applied exceeds the actual variations in the data you’re inevitably obscuring the actual changes that are going on.
The statistics are probably meaningless.
BTW, I’m not sure who said it recently (somebody working at Google?) but I don’t believe that statistics will ever be sexy. Useful, yes. Sexy, no.
I’ve noted before. Be careful of data integrity.
In addition: Q2 GDP benefitted from two transitory items – foreign trade and a 16% (annualized) increase in defense spending. Note that despite considerable (relatively) govt transfers consumer spending was down 1.2%. Ugh.
C4C has done nothing but rob Q4 auto expenditures for Q3. Q3 GDP may now give a false sense of optimism. But loan statistics remain poor, final sales, industrial production and capacity utilization and personal income all still stink.
MSNBC and the networks can cheerlead all they want. There are still great hurdles.
Yeah, that’s basically my take on it, too. The one thing I’ve been pleasantly surprised about in the whole auto sector mess is how well Ford has been doing lately.