The Estimates menu provides additional detail about model parameters. To better understand estimates, you might want to review how JMP codes nominal and ordinal effects. See Details of Custom Test Example, Nominal Factors in the Statistical Details section, and Ordinal Factors in the Statistical Details section).
If your model contains random effects, then only the options below that are appropriate are available from the Estimates menu.
The Estimates menu provides the following options:
Show Prediction Expression
Shows or hides the Prediction Expression report, which contains the equation for the estimated model. See Show Prediction Expression for an example.
Sorted Estimates
Shows or hides the Sorted Parameter Estimates report, which can be useful in screening situations. If the design is not saturated, this report is the Parameter Estimates report with the terms, other than the Intercept, sorted in decreasing order of significance. If the design is saturated, then Pseudo t tests are provided. See Sorted Estimates.
Expanded Estimates
Shows or hides the Expanded Estimates report, which expands the Parameter Estimates report by giving parameter estimates for all levels of nominal effects. See Expanded Estimates.
Indicator Parameterization Estimates
(Available only when there are nominal columns among the model effects.) Shows or hides the Indicator Function Parameterization report, which contains parameter estimates with the nominal effects in the model parametrized using the classical indicator functions. See Indicator Parameterization Estimates.
Sequential Tests
Shows or hides the Sequential (Type 1) Tests report that contains the sums of squares as effects are added to the model sequentially. Conducts F tests based on the sequential sums of squares. See Sequential Tests.
Custom Test
Enables you to test a custom hypothesis. See Custom Test.
Multiple Comparisons
Enables you to specify comparisons among effect levels. These comparisons can involve a single effect or you can define flexible custom comparisons. You can compare to the overall mean, to a control mean, or you can obtain all pairwise comparisons using Tukey HSD or Student’s t. When you specify the Student’s t method, you can also perform equivalence tests to identify pairwise differences that are of practical importance. See Multiple Comparisons.
Compare Slopes
(Available only when there is one nominal term, one continuous term, and their interaction effect for the fixed effects.) Produces a report that enables you to compare the slopes of each level of the interaction effect in an analysis of covariance (ANCOVA) model. See Compare Slopes.
Joint Factor Tests
(Available only when the model contains interactions.) For each main effect in the model, shows or hides a joint test on all of the parameters involving that main effect. See Joint Factor Tests.
Inverse Prediction
Enables you to predict values of explanatory variables for one or more values of the response. See Inverse Prediction.
Cox Mixtures
(Available only when the model contains mixture effects.) Produces parameter estimates for the Cox mixture model. Using these to derive factor effects and estimate the response surface shape relative to a reference point in the design space. See Cox Mixtures.
Parameter Power
Adds columns to the Parameter Estimates report that give power and other details relating to the corresponding hypothesis tests. See Parameter Power.
Correlation of Estimates
Shows or hides a correlation matrix for all parameter estimates in the model. See Correlation of Estimates.
Error Specification
(Available only when there are no random effects.) Specifies the error variance and the error degrees of freedom that are used for standard errors and tests in the Fit Least Squares report. Note that the Studentized Residuals plot and the Box Cox Transformations report are not affected by changing the Error Specification. When the Error Specification is Pure Error or Specified, an additional column appears in the Analysis of Variance report. See Analysis of Variance.
Default Estimate
Uses the standard root mean square error and error degrees of freedom from the model to calculate all tests and standard errors.
Pure Error
Uses the Pure Error mean square and associated degrees of freedom from the Lack of Fit report to calculate all tests and standard errors. See Lack of Fit.
Caution: If the pure error degrees of freedom is 1, a warning message is displayed indicating that tests are weak and confidence limits are large.
Specified
Uses user-specified values for the error variance and error degrees of freedom to calculate all tests and standard errors.