"The alleged margin of error on these reports is one percentage point, so if there is not a big jump in Nowcast on Friday, at least one of the reports is going to be well out of bounds." Really? What if GDP comes in at 3.5%? GDPNow is off by 1% and NY Fed is off by 0.6%. Seems like there's a wiggle room in there of between 3.5 to 3.9. In fact, if the margin of errors are correct, then there is a 90% chance the actual number is in that range.

Government spending is up 5.6%! On what?

Realist said: Government spending is up 5.6%! On what?

Investigating Trump?

Realist said: Government spending is up 5.6%! On what?

Much of the increase was tied to military spending. Also, funds that were previously not spent due to the debt ceiling were released after they repealed or increased the farcical "ceiling" again.

shamrock said: "The alleged margin of error on these reports is one percentage point, so if there is not a big jump in Nowcast on Friday, at least one of the reports is going to be well out of bounds." Really? What if GDP comes in at 3.5%? GDPNow is off by 1% and NY Fed is off by 0.6%. Seems like there's a wiggle room in there of between 3.5 to 3.9. In fact, if the margin of errors are correct, then there is a 90% chance the actual number is in that range.

If you split the difference exactly, you get GDP at 3.65%. That gives you a margin of error of 0.15 PP. I do not know how to calculate that but it is not anywhere close to 90%. That said, and as I pointed out we have yet to hear the Nowcast forecast which comes out Friday. I do expect convergence.

Mike Mish Shedlock said: If you split the difference exactly, you get GDP at 3.65%. That gives you a margin of error of 0.15 PP. I do not know how to calculate that but it is not anywhere close to 90%. That said, and as I pointed out we have yet to hear the Nowcast forecast which comes out Friday. I do expect convergence.

Margin of error is usually quoted at a 95% confidence level. Meaning there is a 95% chance that the actual value of the thing you are estimating is in the range. For example, for GDPNow the range is 4.5% +- 1.0%, or 3.5-5.5. So the odds of GDPNow being in the the margin of error is 95%, and the odds of Nowcast is also 95%. The odds of BOTH being in their respective margin of errors is (0.95 * 0.95) 90.2%. Therefore, there is a 90% probability that the actual GDP is in the overlap of 3.5% to 3.8%.