285508
Rising gas prices: Effect on motorcyclist and bicyclist traffic deaths
Tuesday, November 5, 2013
Andreas Muller, PhD,
Department of Health Policy and Management, UAMS, Fay W. Boozman College of Public Health, Little Rock AR, AR
Purpose: Gas prices have substantially risen and widely fluctuated in recent years. High gas prices tend to reduce motor vehicle travel and traffic deaths. High gas prices may also shift travel to fuel saving, yet riskier modes of transportation. The effect of rising gas prices on motorcyclist and bicyclist traffic deaths is estimated. Data and Method: The time series analysis separately examines national monthly motorcyclist and bicyclist death counts reported by FARS, 1994-12/20011 (n=216). The unobserved component models (UCM) include four deterministic predictors: (1) real average retail gasoline prices, (2) number of new building permits, (3) average surface temperatures, and (4) average precipitation amounts. UCM models allow stochastic representation of time series components (level, slope, season). Using SAS 9.3 software, the UCM models are estimated for 204 months (1/1994-12/2010), validated (1/2011-12/2011), and forecasted for 2012. Results: The analysis estimates that a $1 increase in real gasoline prices is followed by 34 more motorcyclist deaths (t=3.68; p <.001) the next month or about 408 per year. For bicyclists the corresponding estimate is 7 more deaths (t=2.68; p<.01) or about 84 per year. The final time series models fit the fatality counts well both within and out-of-sample (RSQadj.=.966 motorcyclists; RSQadj.=.729 for bicyclists). Model residuals conform to white noise properties. The final models predict well the substantial fluctuations in motorcyclist and bicyclist traffic deaths since 2008. Conclusion: Rising gasoline prices shift highway travel to more fuel saving transportation resulting in increased motorcyclist and bicyclist traffic deaths.
Learning Areas:
Biostatistics, economics
Learning Objectives:
Describe the characteristic features of motorcyclist and bicyclist traffic deaths and predictor time series.
Discuss the unobserved component modeling strategy.
Compare gas price effects for models including all traffic deaths, motorcyclist traffic deaths and bicyclist traffic deaths and present forecasts for the year 2012.
Keyword(s): Motor Vehicles, Statistics
Presenting author's disclosure statement:Qualified on the content I am responsible for because: I have done time series related research since 1990, published time series related analyses involving motor vehicle crashes involving ARIMA models, teach regression based forecasting methods regularly.
Any relevant financial relationships? No
I agree to comply with the American Public Health Association Conflict of Interest and Commercial Support Guidelines,
and to disclose to the participants any off-label or experimental uses of a commercial product or service discussed
in my presentation.
