Dongfeng Chang
;
Apostolos Serletis

the demand for gasoline: evidence from household survey data (replication data)

In this paper we investigate the demand for gasoline in Canada using recent annual expenditure data from the Canadian Survey of Household Spending, over a 13-year period from 1997 to 2009, on three expenditure categories in the transportation sector: gasoline, local transportation, and intercity transportation. In doing so, we use three of the most widely used locally flexible functional forms, the Almost Ideal Demand System (AIDS) of Deaton and Muellbauer (1980), the quadratic AIDS (QUAIDS) of Banks et al. (1997)?an extension of the simple AIDS model that can generate quadratic Engel curves-and the Minflex Laurent model of Barnett (1983), which can also generate quadratic Engel curves. We pay explicit attention to economic regularity, argue that unless regularity is attained by luck, flexible functional forms should always be estimated subject to regularity as suggested by Barnett (2002), and impose local curvature to produce inference consistent with neoclassical microeconomic theory. Our findings indicate that the curvature-constrained Minflex Laurent model is the only model that is able to provide theoretically consistent estimates of the Canadian demand for gasoline. Our estimates show that the own-price elasticity for gasoline demand in Canada is between −0.738 and −0.570 less elastic than previously reported in the literature.

Data and Resources

Suggested Citation

Chang, Dongfeng; Serletis, Apostolos (2014): THE DEMAND FOR GASOLINE: EVIDENCE FROM HOUSEHOLD SURVEY DATA (replication data). Version: 1. Journal of Applied Econometrics. Dataset. http://dx.doi.org/10.15456/jae.2022321.0713404010