The yield is given by the regression equation
y = 859 + 5.76x₁ + 3.82x₂
where
x₁ = number of acres planted
x₂ = number of acres harvested
The goodness of fit is r² = 0.94.
This appears to a very good fit to the data because it is almost equal to 1.
To assess the goodness of fit in a statistical sense, it may also necessary to perform an F-test in a hypothesis test. This is not possible without having raw measured data.
For this problem, r²=0.94 may be considered to be a very good fit to the measured data.
Part a.
When x₁ = 3200 acres and x₂ = 3000 acres, obtain
y = 859 + 5.76*3200 + 3.82*3000
= 30,751 pounds
Part b.
Without performing a hypothesis test or a residual plot, we can conclude that the predicted value is in very good agreement with the actual value.
Because we do not have raw measured data, we can neither plot the residuals nor perform a hypothesis test.
In general,
When r² = 1, the agreement is exact.
When r² = 0, there is absolutely no agreement.
A value of r² > 0.9 is considered good.