by Calculated Risk on 6/11/2005 01:11:00 AM

This month I started to build a simple model to project the trade deficit. I didn't make as much progress as I had hoped, but the first two components (oil imports and China) were somewhat close.

First, I projected oil imports. And then I projected the trade deficit with China.

My model projected $19.3 B Not Seasonally Adjusted (NSA) in energy related petroleum product imports. The actual number was $18.94 B (see Exhibit 17). This is an error of just under 2%.

For the trade balance with China, my model projected a deficit of $15.1B. The actual number (see Exhibit 14) was $14.7B or an error of 2.7%.

Here are each of the components and how the model performed:

ITEM	Projection	Actual	Error
US Exports to China	$3.4B	$3.4B	0
US Imports from China	$18.5B	$18.12B	2.1%
US Trade Deficit: China	$15.1B	$14.7B	2.7%
Oil: Contract Price BBL	$45.70	$44.76	2.1%
Oil: BBLs Crude	328.6M	313.8M	4.7%
Oil: Price Other BBL	$52.56	$50.77	3.5%
Oil: BBLs Other	82M	96.5M	15%
ERPP Total NSA	$19.3B	$18.94B	1.9%

UPDATE: I found an error in the oil model. I used 31 days for April to estimate the quantity of Crude. The actual Q(crude) should have been 318 million BBL, or an error of 1.3% (instead of 4.7%). The overall oil imports projection should have been $18.83B vs. actual of $18.94B. That is a dumb error!

I didn't have a good method for estimating "Other" energy related petroleum products. As I wrote: "The good news is that the larger percentage errors for "Other" are not very important for the overall ERPP."

There is much more to do!