A dynamic factor model of the coincident indicators for the US transportation sector

Abstract

This paper studies the business cycle features of the transportation sector using dynamic factor models. The transportation reference cycles peak ahead of the economic cycles, but lag by a few months at troughs. The asymmetric relationship between these two suggests the usefulness of transportation in monitoring business cycles.

Acknowledgements

This study has been supported by grants from the Bureau of Transportation Statistics, US Department of Transportation during 2001–2003. However, the contents of this paper reflect the views of the authors, who are responsible for the facts and the accuracy of the information presented herein.

Notes

 Gordon (1992) and Bosworth (2001) have provided valuable insights into the different methodologies and data that BEA and BLS use to construct alternative annual transportation output series. A comparison suggests that these annual output measures reflect the long-term trends of TSI, and that the latter is superior in reflecting the cyclical movements in the transportation sector.

 This nonparametric approach includes four steps: (1) month-to-month changes (x _t ) are computed for each component (X _t ) using the conventional formula; (2) the month-to-month changes are adjusted to equalize the volatility of each component using the standardization factors; (3) the level of the index is computed using the symmetric percentage change formula; and (4) the index is re-based to be 100 in 1996 to make a formal NBER index.

 Interestingly, a similar asymmetry also exists between inventory and business cycles, see Zarnowitz (1992, p. 336) and Humphreys et al. (2001).