*
* F-test examples on page 137
*
open data andy.dat
data(format=free,org=columns) 1 75 sales price ads
*
linreg sales
# constant price ads
*
exclude
# price
exclude
# price ads
*
* SET is the instruction for doing general transformations of full data series.
* This creates ASQ as the square of the ADS series.
*
set asq = ads^2
linreg sales
# constant price ads asq
*
exclude(title="Test of Advertising Variables")
# ads asq
*
disp "Optimal Ad Expenditure" (1-%beta(3))/(2*%beta(4))
*
* RESTRICT is the most general of the hypothesis testing instructions. It can test
* hypotheses involving linear combinations of the coefficients. Each restriction
* is represented by a pair of # lines. The first has the coefficient positions
* entering the restriction, the second has the corresponding multipliers followed
* by the target value for that linear combination. In this case, we're testing
*
* 1.0 x B(3) + 3.8 x B(4) = 1.0
*
* The wizard for RESTRICT is "Regression Tests" on the "Statistics" menu. Choose
* "General Linear Restrictions".
*
restrict(title="Test of Optimal Expenditure") 1
# 3 4
# 1.0 3.8 1.0
*
* Joint hypothesis on page 145
*
restrict(title="Joint Hypothesis") 2
# 3 4
# 1.0 3.8 1.0
# 1 2 3 4
# 1.0 6.0 1.9 1.9^2 80.0