Doing prediction scenarios with the regression model
As we saw earlier, the regression model is not an exact answer for our prediction's needs because it depends on the level of the relationship between the causal variables and the Y result/effect variable.
Taking this into account, a more realistic answer is to present three scenarios of the prediction model. The scenarios depend on the confidence level of the slope. We have the upper and lower scenarios, depending on the value assigned to the slope. These scenarios exhibit the different value range for miles-per-gallon performance, depending on the variation of horsepower from 0 to 150:
- Linear model scenario: Miles-per-gallon performance goes from 34 to 14
- Upper scenario: Performance of miles per gallon has a range between 30 and 17
- Lower scenario: Performance of miles per gallon goes from 30 to 0
The formula for the confidence level of the slope is as follows:
From previous...
