|
Poisson RegressionRegression Equation |
The regression equation is an algebraic representation that describes the relationship between the response and predictor variables. The form of the regression equation with respect to the number of occurrences depends on the link function. Use the equation to predict the number of occurrences.
Minitab provides a separate regression equation for each level of each categorical predictor in the model.
Example Output |
Regression Equation
Discoloration defects = exp(Y')
Size of Screw large Y' = 4.398 + 0.01798 Hours Since Cleanse - 0.001974 Temperature
small Y' = 4.244 + 0.01798 Hours Since Cleanse - 0.001974 Temperature |
Interpretation |
|
For the resin defect data, there are two equations because the size of the screw is a categorical variable with 2 levels. Because there is no interaction in the model, the coefficients for the continuous predictors are the same in both equations.
In the absence of interactions, you can evaluate the relative number of defects with a comparison of the constants in the equations. For the resin defect data, the constant in the equation for the small screw is the least (4.244). The smaller constant means that the smaller screw averages fewer defects than the large screw.