Estimating traffic demand risk – A multiscale analysis

This paper proposes a novel method for estimating the traffic demand risk associated with transportation. Using mathematical properties of wavelets, we develop a statistical measure of traffic demand sensitivity with respect to GDP. This measure can be adapted in a flexible way to capture risk levels relevant for different investment horizons. We demonstrate the timescale decomposition of risk with Swedish traffic demand data for 1950–2005. In general, rail transport shows a stronger co-movement with GDP than road transport. Moreover, we examine the volatility exhibited by traffic demand. Our findings suggest that rail investments are more risky than road investments. Since the findings can be used for optimal investment timing and for choice between public investment alternatives, they are deemed important for public policy in general.

► Traffic demand variance and GDP-sensitivity decrease with length of time scale.
► For most time scales traffic demand on railroad is more GDP-sensitive than for road.
► Freight transportation is generally more GDP-sensitive than passenger transportation.
► The pattern of time-scale dependent volatility affects optimal investment timing.
► GDP-sensitivity affects choice among investments via risk adjusted discount rate.