Estimation of the Tax Rates Based on Vehicle Miles Traveled Using Stochastic Models

摘要

In this thesis, we shall study the alternative revenue collection system which is based on the vehicle miles traveled (VMT). In various studies, it has been found that the existing revenue collection system based on gas/fuel tax is not an appropriate option in the longer run. The main reasons include no effective tax process for vehicles based on alternative fuel vehicle, no effective changes in tax due to economical inflation, and more highway expenditure than generated revenue. Our main objective is to estimate the VMT tax rates that should be charged in order to generate same amount of revenue generated by gas tax. It is apparent that the amount of gas consumed is dependent on the behavior of gas prices which fluctuate daily. Also, VMT is dependent upon the amount of gas consumed and thus it is also dependent on the gas prices. Different mathematical models based on stochastic deferential equations shall be developed for gas prices, the amount of gas consumed, and VMT. Parameters for all the proposed models shall be estimated by using maximum likelihood principle technique and the historical data. As result of our simulation, we have found that VMT tax rate should be approximately 2.5 cents per mile in order to generate same amount of revenue as generated by current system. This VMT tax rate is close enough to the estimated value of 2 cents per mile by Nevada Department of Transportation