Fitting Linear Models • Model Specification • Model Specification Options
Previous
 • 
Next
Model Specification Options
Descriptions of the Model Specification Options describes the options in the red triangle menu next to Model Specification.
Descriptions of the Model Specification Options 
Center Polynomials
Causes any continuous term involved in an effect with degree greater than one to be centered by its mean. This option is checked by default, except when a term involved in the effect is assigned the Mixture Effect attribute or has the Mixture column property. Terms with the Coding column property are centered midway between their specified High and Low values.
Centering is useful is making regression coefficients more interpretable and in reducing collinearity between model effects.
Informative Missing
Selecting this option provides a coding system for missing values. This system allows estimation of a predictive model despite the presence of missing values. It is useful in situations where missing data are informative. See Informative Missing for more details.
This option is available for the following personalities: Standard Least Squares, Stepwise, Generalized Regression, MANOVA, Loglinear Variance, Nominal Logistic, Ordinal Logistic, Proportional Hazard, Parametric Survival, Generalized Linear Model, and Response Screening.
Set Alpha Level
Sets the alpha level for confidence intervals in the Fit Model analysis. The default alpha level is 0.05.
Save to Data Table
Saves your Fit Model launch window specifications as a script that is attached to the data table. The script is named Model. When a table contains a script called Model, this script automatically populates the launch window when you select Analyze > Fit Model. (Simply rename the script if this is not desirable.)
For details about JSL scripting, see the JMP Scripting Guide.
Save to Script Window
Copies your Fit Model launch window specifications to a script window. You can save the script window and re-create the model at any time by running the script.
Create SAS job
 
Creates a SAS program that can re-create the current analysis and data table in SAS in a script window. Once created, you have several options for submitting the code to SAS.
1.
Copy and paste the code into the SAS Program Editor. This method is useful if you are running an older version of SAS (pre-version 8.2).
2.
Select Edit > Submit to SAS.
3.
Save the file and double-click it to open it in a local copy of SAS. This method is useful if you would like to take advantage of SAS ODS options, such as generating HTML or PDF output from the SAS code.
See the Using JMP book.
Submit to SAS
 
Submits code to SAS and displays the results in JMP. If you are not connected to a SAS server, prompts guide you through the connection process.
See the Using JMP book.
Convergence Settings
Note: This option appears only for certain personalities or if REML is selected as the method in the launch window.
Depending on the model choice, the Convergence Settings menu includes the following options:
•
Maximum Iterations
•
Convergence Limit
Default values for the Maximum Iterations and Convergence Limit are 100 and 0.00000001, respectively. You can click on these values to enter new values for these limits. If you have a very large data set or a complicated model, you might want to limit the number of iterations. If your model does not readily converge, you might want to either increase the Maximum Iterations or increase the Convergence Limit.
Informative Missing
The Informative Missing option constructs a coding system that allows estimation of a predictive model despite the presence of missing values. It codes both continuous and categorical model effects.
Continuous Effects
When a continuous main effect has missing values, a new design matrix column is created. This column is an indicator variable, with values of one if the main effect column is missing and zero if it is not missing. In addition, missing values for the continuous main effect are replaced with the mean of the non-missing values for rows included in the analysis. The mean is a neutral value that maintains the interpretability of parameter estimates.
The parameter associated with the indicator variable estimates the difference between the response predicted by the missing value grouping and the predicted response if the covariate is set at its mean.
For a higher-order effect, missing values in the covariates are replaced by the covariate means. This makes the higher-order effect zero for rows with missing values, assuming that Center Polynomials is checked (the default setting). This is because Center Polynomials centers the individual terms involved in a polynomial by their means.
In the Effect Tests report, each continuous main effect with missing values will have Nparm = 2. In the Parameter Estimates report, the parameter for a continuous main effect with missing values is labeled <colname> Or Mean if Missing and the indicator parameter is labeled <colname> Is Missing. Prediction formulas that you save to the data table are given in terms of expressions corresponding to these model parameters.
Categorical Effects
When a nominal or ordinal main effect has missing values, the missing values are coded as a separate level of that effect. As such, in the Effect Tests report, each categorical main effect with missing values will have one additional parameter.
In the Parameter Estimates report, the parameter for a nominal effect is labeled <colname>[ ]. For an ordinal effect, the parameter is labeled <colname>[-x], where x denotes the level with highest value ordering.
As with continuous effects, prediction formulas that you save to the data table are given in terms of expressions corresponding to the model parameters.
Coding Table
When you are using the Standard Least Squares personality, you can view the design matrix columns used in the Informative Missing model by selecting Save Columns > Save Coding Table.