Home / DOE / Additional Information / Types of Outputs
Types of Outputs¶
An output's Type: decides which regression engine fits it, what the prediction means, and which charts will offer it. The list has exactly three entries.
An output is Quantitative, Binary or Nominal, and each is fitted by a different engine.
A Quantitative output is fitted by least squares and predicted as a value, with a separate model available for its standard deviation.
A Binary output has two outcomes and is fitted by binary logistic regression. The prediction is the probability of the level you selected, not a value on the response scale.
A Nominal output has three or more unordered outcomes and is fitted by nominal logistic regression. The prediction is again a probability, one per level, and the probabilities across the levels sum to one.
The three types, side by side¶
| Quantitative | Binary | Nominal | |
|---|---|---|---|
| Engine | Ordinary Least Squares | Binary Logistic | Nominal Logistic |
| Prediction | a value on the response scale | the probability of the selected level | a probability per level, summing to one |
| Second model for the spread | yes, the S-Hat model | no | no |
| Weights checkbox reads | Has weights column | Events / Sample size | Has weights column |
| Offered by Observed vs Predicted | yes, as Y-Hat or S-Hat | no | no |
| Offered by Residual Plots | yes | yes | no |
| Cpk and DPM available | yes, with a specification limit | no | no |
There is no ordinal type¶
The Type: list offers Quantitative, Binary and Nominal, and nothing else. Quantum XL has no ordinal logistic regression, so an ordered response such as a 1 to 5 rating is entered as one of the three above.
Binary outputs and the events form¶
A binary output can be entered two ways. One row per unit, with the outcome in the response column, or one row per design point with Events / Sample size ticked, which adds a pair of columns for the number of events and the number of trials.
The events form is available only when the design format is Stacked. See Stacked vs. Table.
Mixed workbooks¶
A design can carry outputs of different types at once, and each is fitted by its own engine in a single run of the regression. Each gets its own regression area on the regression sheet, and the charts offer whichever output levels apply to them.