| By applying market-level data on quantities, prices and vehicle characteristics in the year of 2010 and 2013, I conduct an estimation of automobile demand in the U.S. market using the conditional logit and nested logit model. By testing and applying the instruments variables, I address the exogeneity problem of prices and within-group market share. The demand for cars and light trucks are both elastic. The nested logit regression indicates that models within the same group/nest are better substitutes than models across the group. The results imply that consumers prefer the efficient vehicles. However, the electric vehicles and low displacement vehicles are not preferred over high displacement fuel vehicles by consumers. |
