A study of cigarette demand resulted in the following logarithmic regression equation:

log(Q) = -2.55 - .29 log(P) - .09 log (Y) + .08 log(A) - .1W

(-2.07)        (-1.05)       (4.48)          (-5.2)

Here, Q denotes annual cigarette consuption, P is the average price of cigarettes, Y is the per capita income, A is total spending on cigarette advertising, and W is a dummy variable whole value is 1 for yeas after 1963 and 0 for ealier years.  The t-statistic for each coefficiient is shown in parenthesis.  The R2 of the equation is .94.

a) What does the coeffiecient of log(P) represent?  If cigarette prices increase by 20 percent, how will this affect consumption?

b) Are cigarette purchases sensitive to income?  Explain