Question

. XI: reg InGas i.month InP Ininc i·month if date>=190&date«-250 _Imonth_1-12 (naturally coded; -1month-1 omitted) Source df MS Number of obs - 61 F(13, 47) Model Residual 284473817 13 .021882601 Prob >F 47 0 R-squared 1.0000 Adj R-squared1.0000 Total 284473817 60 .00474123 Root MSE lnGas Coef. Std. Err. [95% Conf. Interval] lnp lnInc Imonth 2 Imonth_3 Imonth 4 Imonth 5 Imonth_6 Imonth_7 Imonth_8 Imonth_9 -1month-10 Imonth 11 _Imonth_12 cons 1 1.14e-16 1.48e-17 -3.27e-17 -2.67e-17 -4.46e-17 -4.55e-17 -4.05e-17 -4.34e-17 -2.04e-17 -3.39e-17 -2.07e-17 -3.92e-17 -2.66e-15

Interpret the coefficient on logged real gasoline price (lnP) in terms of the sign, magnitude and statistical significance. What does this estimate tell us about the average response of gasoline demand to changes in prices from 1975-1980.

[lnGas = ln(gascap)

lnP = ln(realprice)

lnInc = ln(inccap)

realprice = Real price of gasoline (2000 $)

gascap = Gas demanded per capita (gallons per month)

inccap = Real income per capita (2000 $)

date = Year and month of observation (text)

year = Year of observation

month = Month of observation]

Answer 1

As both independent and dependent variables are in log form, the coefficient could be interpreted in elasticity term.

In this case, as real price of gasoline increases by 1%, the demand of the gasoline will increase by 1%. (The answer is based on the regression output shown in the question.)

As per the model, income elasticity of demand of gasoline is less than 1.

P.S. I doubt the validity of the regression result posted in the question. Methodology used to analyse the data may not be right. However, I am submitting my answer considering the output is correct.