Frequently Asked Questions
How does ATA compile the TRAC data?
Each month ATA sends surveys out to for-hire motor carriers asking them for the data published in TRAC. ATA then matches up the carriers in the latest month and only uses those carriers that also provided data in the previous month. We then calculate a percent change for all the carriers combined (or for selected groups, like dry van) for all data categories (e.g., loads, revenue, miles, etc) and move the previous not seasonally adjusted index for that category up or down by the percent change.
Why is the TRAC data in index form instead of actual numbers?
Since our data is based on surveys, we do not attempt to say how many loads were transported by all for-hire carriers. Instead, the data should be used to look at trends and percent changes. Several government economic indicators are also in index form, like the Federal Reserve's Industrial Production. ATA's monthly truck tonnage indicator is also an index.
What is seasonally adjusted data?
Our seasonally adjusted indexes have been "adjusted" to remove effects of changes that normally occur around the same time every year (holidays, the fall freight season) as well as the days of the month, which can increase or decrease freight volumes simply because there are more or less days compared to the previous month. Seasonally adjusted data should isolate trends or changes due to the business cycle and not due to reoccurring annual events.
How does ATA seasonally adjust the data?
An adjusted index is obtained by applying statistical seasonal coefficients to the unadjusted index. Those coefficients are built from a statistical process that weighs in past fluctuations in the unadjusted index and attempts to correct for pronounced changes that occur repeatedly due to seasonal phenomena. The goal is to remove the effect of these phenomena and to obtain a better picture of the ongoing business trend. We use software produced by the U. S. Census Bureau called X-12. This is the standard software used by government agencies to seasonally adjust many of its data series.
Why is the total industry index sometimes larger or smaller than both the large and small carrier indexes? Shouldn't it be in between the two?
If the TRAC indexes were a so-called “fixed weighted” index, then the total index should always be in between the indexes comprising the total. And this is always true for the month-to-month percent change in the not seasonally adjusted data. However, since our indexes are not fixed weighted, meaning the relative importance of the sub-groups can change each month, you will find that occasionally the total index level may not be between the sub-component indexes. While this is fairly rare, it does happen. It is also the case that the percent changes in seasonally adjusted indexes can be larger or smaller than changes in the sub-indexes.
What series should motor carriers benchmark to, seasonally or not seasonally adjusted?
For benchmarking purposes (i.e., to compare their data against the industry group(s)), carriers should use not seasonally adjusted indexes because the percent changes are what the carriers actually reported. This is true for month-to-month and year-over-year comparisons. However, to better see the underlying changes in the industry and trends in general, it is best to use the seasonally adjusted data since it strips out effects that occur around the same time each year (e.g., holidays).
Why do the indexes say 2000=100?
The indexes are set to 2000=100 because 2000 is the base year. In general, any year or month of a year (e.g., January 1993) can be made the base year.
Why did ATA choose 2000 as the base year?
We chose 2000 as the base year because it makes for easier comparisons compared to 1993 as well as it was the first year the industry saw a downturn in the economy. This permits us to see how the industry has improved since then.