Statistics for Business: Decision Making and Analysis, 3rd Edition

1. 1. $\mathrm{...\; 814}-2\left(2,930\right)<0$. Negative sales are not possible. The interval is too long.

Fixing the Problem: Revise the Model

Most regression modeling is iterative. No one gets it right the first time, every time. We try our best, see how it works, and then refine the model as we learn from the data.

Let’s think in terms of fixed costs and variable costs. If F represents fixed costs and M marginal costs, the equation of the SRM becomes

$$\text{Price}=F+M\times \text{Sq} \text{Ft}+\epsilon$$

The problem in this example is that the errors ∊ vary more for large homes than those for small homes. The errors in regression accumulate the influences of the response of other variables aside from the single explanatory variables in the model. Evidently other factors that influence the price of a home, ...